Report No. 78318-PY Growth Volatility in Paraguay Sources, Effects, & Options June 5, 2014 Argentina, Paraguay and Uruguay Country Management Unit Poverty Reduction and Economic Management Latin America and the Caribbean Region Document of the World Bank CURRENCY AND EXCHANGE RATE (As of June 5, 2014) CURRENCY UNIT = GUARANIES US$1.00 = 4421 Guaraníes FISCAL YEAR January 1 – December 31 Abbreviations and Acronyms ASERCA Supports and services for MATBA Futures and options exchange in agricultural marketing (Apoyos y Buenos Aires, Argentina servicios a la comercialización (Mercado a termino de Buenos agropecuaria.) Aires S.A.) ARM Agricultural risk management MERCOSUR Southern common market (Mercado Común del Sur) AYII Area-based Yield Index Insurance MEF Ministry of Finance (Ministerio de Hacienda) AxC Contract based agriculture MICs Middle Income Countries (Agricultura por contrato) BCP Central Bank of Paraguay (Banco OECD Organization for Economic Central de Paraguay) Cooperation and Development CADENA Component Attention to Natural PHEFA Hemispheric Plan of Disasters in the Agricultural and Eradication of Foot and Mouth Fisheries (Componente Atencion a Disease Desastres Naturales en el Sector Agropecuario y Pesquero) CBOT Chicago Board to Trade PEMEX Mexico's state-owned petrol company (Petróleos Mexicanos) CSF Chile’s Copper Stabilization Fund PIT Personal income tax CIT Corporate income tax PPP Public-private partnership EPH National Household Survey RENAMU National registry of (Encuesta Permanente de Hogares) municipalities (Registro Nacional de Municipalidades) ii FMD Foot and mouth disease ROFEX Futures and options exchange in Rosario, Argentina (Mercado a termino de Rosario S.A. GAP Good agriculture practices SENACSA National Service of Animal's Quality and Health (Servicio Nacional de Calidad y Salud Animal) GDP Gross Domestic Product SPF Norway’s Stabilization State Petroleum Fund IBLIP Index-Based Livestock Insurance SPS Sanitary and Phytosanitary Program IMAGRO Agricultural corporate income tax VAT Value Added Tax DGEEC National Institute of Statistics WTO World Trade Organization (Dirección Nacional de Encuestas Estadísticas y Censos) ISC Selective Consumption Tax y-o-y Year-on-year (Impuesto Selectivo al Consumo) LAC Latin America and the Caribbean MICs Middle Income Countries Vice President: Jorge Familiar Calderon Country Director: Jesko S. Hentschel Sector Director: J. Humberto Lopez Sector Manager: Auguste Tano Kouame Sector Leader: Zafer Mustafaoglu Task Team Leaders: Friederike (Fritzi) Koehler-Geib iii Contents Abbreviations and Acronyms ......................................................................................................... ii Executive Summary ........................................................................................................................ 2 Chapter 1: The Sources of Volatility in Paraguay ..................................................................... 5 1.1. Stylized facts ................................................................................................................. 6 1.2. Sources of volatility .................................................................................................... 11 1.3. The role of factor markets ........................................................................................... 18 Chapter 2: The effects of growth volatility in Paraguay with a focus on volatility originating in the agricultural sector ................................................................................. 20 2.1 Propagation of shocks within agriculture ................................................................... 20 2.2 The impact of volatility originating in the agricultural sector on other sectors .......... 24 2.3 The impact of volatility originating in the agricultural sector on macroeconomic aggregates .................................................................................................................................. 28 Chapter 3: Policy options for the management of growth volatility in Paraguay ................ 32 3.1 The macroeconomic toolbox to address growth volatility.......................................... 33 3.2 The agricultural risk management toolbox ................................................................. 38 3.3 Combining the macroeconomic and the agricultural risk management toolbox ........ 55 List of Figures Chapter 1 Figure 1.1: Real GDP growth ....................................................................................................................... 6 Figure 1.2: Volatility over time, Paraguay in regional comparison .............................................................. 7 Figure 1.3: GDP—breakpoints of volatility of quarterly y-o-y GDP growth using Inclan, Tiao (1994)...... 8 Figure 1.4: Agricultural GDP—breakpoints of volatility of quarterly y-o-y GDP growth using Inclan, Tiao (1994).................................................................................................................................................... 8 Figure 1.5: Share of agriculture in GDP ....................................................................................................... 9 Figure 1.6: Growth volatility by economic sector ........................................................................................ 9 Figure 1. 7: Volatility of rainfall ................................................................................................................. 10 Figure 1. 8: Rainfall and agriculture GDP .................................................................................................. 10 Figure 1. 9 Exports by product.................................................................................................................... 11 Figure 1.10: Correlation between trade balance and world interest rate ..................................................... 13 Figure 1.11: Impulse response of trade balance to a shock to the world interest rate ................................. 13 Figure 1.12: Correlation between TOT and world interest rate .................................................................. 14 iv Figure 1.14: Business cycle fluctuations in Paraguay—Government investment versus GDP .................. 15 Figure 1.15:Contribution of public and private demand components to real GDP growth ........................ 16 List of Figures Chapter 2 Figure 2.1: Wavelet analysis of rainfall and agricultural GDP ................................................................... 21 Figure 2.2: Impulse response functions linking Paraguay’s agricultural GDP to world............................. 23 Figure 2.3: Impulse response functions linking Paraguay’s agricultural GDP to the construction sector .. 25 Figure 2.4: Impulse response functions linking Paraguay’s agricultural GDP to the services sector......... 26 Figure 2.5: Wavelet analysis of agriculture and non-agricultural GDP ...................................................... 28 Figure 2.6: Wavelet analysis of agriculture and private consumption ........................................................ 29 Figure 2.7: Wavelet analysis of non-agriculture and private consumption................................................. 29 List of Figures Chapter 3 Figure 3.1: The World Bank Agricultural Risk Management Framework ................................................. 39 List of Tables Chapter 1 Table 1.1: Export by destination ................................................................................................................. 11 Table 1.2 Variance decomposition of GDP volatility ................................................................................. 12 Table 1.3 Correlations across variables ...................................................................................................... 14 Table 1.4: Variance Decomposition of Agricultural and Non-Agricultural GDP volatility ....................... 17 List of Tables Chapter 3 Table 3.1: Instruments for Managing Production Risk ............................................................................... 39 Table 3.2 Colombia’s study case. Benefits, Challenges and Considerations for Paraguay ........................ 41 Table 3.3: Mexico’s study case. Benefits, challenges and considerations for Paraguay ............................ 44 Table 3.4: Malawi’s study case. Benefits, challenges and considerations for Paraguay ........................... 45 Table 3.5: Mongolia’s study case. Benefits, challenges and considerations for Paraguay ......................... 47 Table 3.6: Instruments for Managing Market Risk ..................................................................................... 48 Table 3.7: Peru’s study case. Benefits, Challenges and considerations for Paraguay ................................ 49 Table 3.8: Subsidy Components, AxC Program, ASERCA ....................................................................... 52 Table 3.9: Mexico’s study case. Benefits, Challenges and considerations for Paraguay ............................ 52 Table 3.10 : Comparison CBOT, MATBA, & ROFEX.............................................................................. 53 Table 3.11: Argentina’s study case. Benefits, challenges and considerations for Paraguay ....................... 54 List of Annexes v . Annex 1.1: Volatility over time, international comparison......................................................................... 63 Annex 1.2: Volatility breaks of macroeconomic variables in Paraguay ..................................................... 64 Annex 1.3: Graphs on volatility breakpoints Inclan Tiao (1994) by variable............................................. 66 Annex 1.4: Sectoral GDP correlations ........................................................................................................ 70 Annex 3.1: Traditional measures for agricultural risk management ........................................................... 72 Annex 3.2: Insurance products.................................................................................................................... 72 vi This report was prepared by a team led by Friederike (Fritzi) Koehler-Geib (LCSPE) under the overall supervision and guidance of Zafer Mustafaoglu (Lead Economist and Sector Leader, LCSPR), Auguste T. Kouame (Sector Manager, LCSPE), J. Humberto Lopez (Sector Director, LCSPR), Rodrigo A. Chaves (former Sector Director, LCSPR) and C. Penelope Brook (Country Director, LCC7C). The peer reviewers were Aristomene Varoudakis (Advisor, DECOS), Cesar Calderon (Senior Economist, DECWD), Julie Dana (Lead Financial Officer, FABLO), and Norbert Fiess (Principal Economist/Credit Risk Head, CFRCR). The core team included Elida Caballero Cabrera, Diana Lachy, Rei Odawara, Guillermo Cabral, Jorge Araujo, Miriam Beatriz Villarroel, Marcelo Echague, Patricia Chacon Holt, Peter Siegenthaler, Silvia Gulino (all LCSPE), Dante Mossi, (Country Manager, LCCPY), Gloria Dure, Rosa Arestivo de Cuentas Zavala, (all LCCPY), and Rossana Polastri (former Country Manager, LCCPY). Inputs and background papers were also received from Sophie Storm Theis (LCSSD), Diego Arias Carballo (LCSAR), Hakan Berument, Julio Ramirez, Viktoria Hnatkovska (all consultants), Andres Lajer Baron, Carolina Saizar, Hannah Nielsen, Nathalie Picarelli, Pia Maria Zanetti, and Sona Varma (all LCSPE), Oscar Calvo-Gonzalez (LCSPR), Julian Lampietti (LCSSD), David Gould (SARCE). Comments and inputs were also received from many colleagues working in the Paraguay country team, including Andrew Follmer, Carla Cutolo, Elena Feeney, Mariela Alvarez, Sabine Hader (all LCC7C). The team is thankful for the excellent collaboration with the Ministry of Finance, in particular with the vice ministry of economics including the departments of economic studies, Macro Fiscal Policies, and Debt Policy. 1 Paraguay’s real GDP growth has been one of the most volatile in the region in recent years. Between 2000 and 2011, real GDP growth in Paraguay fluctuated by 5.5 percentage points, exceeding the volatility of most Latin American peer countries. This was not always the case. During the period 1960-2011, growth volatility in Paraguay was lower than in other countries in the region. It is too early to tell whether high volatility in Paraguay is temporary or permanent, even though some structural changes as for example the increase in the weight of the agricultural sector in GDP are in line with the idea that volatility is there to stay. The high level of volatility is concerning because of the significant costs associated with it in terms of welfare, economic growth, and equality. For developing countries, macroeconomic volatility, summarized by output volatility, is reflected disproportionately in consumption volatility, and welfare gains from reducing consumption volatility can be substantial (Loayza, Ranciere, Serven, and Ventura (2007)). No less important is the negative impact of volatility on economic growth. The impact arises through a decrease in productivity and various forms of uncertainty such as economic, political, policy-related, as well as a tightening of binding investment constraints.1 The negative link between macroeconomic volatility and equality has also been established in the literature.2 Designing policies that help mitigate the impact of shocks to the economy and that help increase the country’s resilience is particularly relevant in this light, also because Paraguay still has a low per capita income compared to its neighbors and continues to suffer from a high degree of inequity and poverty. The high volatility of GDP growth has coincided with a volatile macroeconomic environment. A large number of relevant economic variables and variables with economic significance have shown high levels of volatility in recent years, including the world interest rate, Paraguay’s nominal exchange rate, its current account balance, public consumption and investment, credit to the private sector, agricultural GDP, rainfall and, soy prices. While the agricultural sector was particularly affected, preceded by an increase in the volatility of soy prices and rainfall, non-agricultural GDP actually registered a decrease in volatility. External shocks explain over 50 percent of Paraguay’s GDP growth volatility. A key factor behind volatility in Paraguay is the strong dependence on agriculture and its concentration on few products and few export destinations, both of which have increased over time. Of the external shocks, foreign demand for Paraguayan output accounts for about 30 percent of GDP volatility, the world interest rate for 20 percent, and terms of trade for 3 percent. The impact of the world interest rate runs through its impact on portfolio reallocation, commodity prices, and economic conditions of main trading partners. Fluctuations in commodity prices and world real interest rates have hit all countries in the region and the world. However, their output response reflects the interaction of these shocks with country-specific conditions that range from the strong dependence on a few goods and services or a narrow tax base and economic policies. 1 A large body of literature has addressed this topic from various perspectives Acemoglu et al (2003), Aizenman and Pinto (2005), Berument, Dincer, and Mustafaoglu (2011), Ramey and Ramey (1995) and Wolf (2005). 2 See for example Breen and Garcia-Penalosa (2004), Garcia-Penalosa and Turnovsky (2004) or Huang, Fang, and Miller (2012). 2 Domestic variables explain the remaining share. 25 percent of stemming from shocks to real GDP, 15 percent from shocks to investment, and 3 percent from pro-cyclical fiscal and monetary policy. Pro-cyclicality is defined as a positive response of government spending to an exogenous expansionary business cycle shock. In developing countries it is usually linked to limited access to credit in downturns, to lax fiscal stances in good times, to and burdensome bureaucratic processes. While Paraguay shares this pattern there is an indication that the fiscal stance was counter-cyclical during the contractions in 2009 and 2012. In contrast to overall GDP volatility, fluctuations of agricultural GDP originate to a large extent from domestic shocks, with weather related shocks to agricultural output itself accounting for more than a third. Despite a significant decrease, persistent rigidities in factor markets and limited mobility across sectors reduce the economy’s capacity to buffer shocks and exacerbate business cycle fluctuations and hence volatility. While labor market distortions have declined, firms’ access to credit have improved, and agricultural efficiency has increased, important challenges remain which suggest that shocks hitting Paraguay may rather be exacerbated than buffered. In particular, labor and capital returns between agriculture and non-agriculture remain large, suggesting limited factor mobility across sectors, financing constraints facing households have remained pronounced and time-varying and the efficiency in the non-agricultural sector has shown no signs of improvement, to the contrary has been deteriorating. These remaining frictions reduce efficiency of the Paraguayan economy and prevent its capacity to buffer shocks that hit the economy. Growth volatility impacts Paraguay’s economy in various ways, it impacts : i) the agricultural sector; ii) other sectors; and iii) macroeconomic aggregates such as investment, tax revenues, or poverty and equity. The main sources of volatility in the agricultural sector can be categorized into shocks to production and shocks to markets. Shocks to production include variations in rainfall, investment levels, and disease outbreaks (e.g., foot and mouth disease). Shocks to markets include commodity price variations, the closing of markets in the case of disease outbreaks, and fluctuations of prices of imported inputs like fertilizers and pesticides. Overall, market participants in Paraguay report that a lack of information and knowledge on the patterns and impacts of volatility on the economy is the biggest challenge to operating within this environment. Within the agricultural sector sources of volatility manifest themselves through shocks that impact the level of infrastructure and R&D investment; cause payment delays; and trigger the use of diversification strategies. In terms of other sectors, the volatility of agricultural GDP mainly affects those economic activities that provide inputs such as machinery or storage and transport services, but it also affects the financial services and construction sectors. In terms of macroeconomic aggregates, the exchange rate and levels of employment fluctuate as a consequence of shifts in agricultural exports. There is some indication that private consumption plays a role in propagating the impact of agricultural GDP through the economy, impacting non-agricultural GDP. Investment levels are lower, fiscal revenues are indirectly affected, and the reduction of poverty and inequity are generally slower than in countries with lower levels of volatility. In terms of managing volatility, it is important to develop a comprehensive macroeconomic risk management framework that takes all different sources of volatility and risks into account. Sources of volatility are interrelated and taking a broader perspective allows finding optimal ways to manage observed volatility and risks. Any policy option needs to be assessed in 3 terms of its fiscal implications; be it in terms of its effects on sustainability, on redistribution, or on potential contingent liabilities. Policy options are presented as: i) a macroeconomic tool set; and ii) an agricultural risk management tool set, which need to be aligned with one another in an overall framework. Macroeconomic policy options include the development of a strategy on the role of agriculture in the economy and its structure; policies that render factor markets flexible; and fiscal policies such as the introduction of fiscal rules and stabilization funds. The agricultural risk management tool box is designed to address the shocks to production and to markets specific to the agricultural sector. First, four case studies are presented on new tools and approaches to mitigate, cope with, and transfer agricultural production risks: i) building animal health capacity to prevent foot and mouth disease in Colombia; ii) introducing weather derivatives based on a rainfall index for severe drought in Malawi; iii) establishing a weather contingency fund for the agricultural sector (CADENA) in Mexico; and iv) implementing an index-based livestock insurance project in Mongolia. Second, three case studies provide examples of measures to mitigate and transfer agricultural market risks: i) developing the asparagus market in Peru; ii) introducing subsidies for commodity price hedging contracts in Mexico; and iii) introducing agricultural commodity exchanges in Argentina. While all case studies have been selected based on their relevance for Paraguay, a careful assessment of their applicability to Paraguay would be required as part of an overall assessment of agricultural risks. Government and the World Bank have been engaging in a dialogue on this topic through the preparation of this study and with a joint agricultural risk management assessment. 4 Chapter 1: The Sources of Volatility in Paraguay Paraguay’s real GDP growth has been one of the most volatile in the region in recent years. This was not always the case. During the period 1960-2011, growth volatility in Paraguay was lower than in other countries in the region and many of these countries managed to reduce volatility (see Table Annex 1). The high level of volatility in Paraguay is concerning because of the significant costs associated with it in terms of welfare, economic growth, and equality. 3 Designing policies that help mitigate the impact of shocks to the economy and that increase the country’s resilience is particularly relevant in this light, and also because Paraguay still has a low per capita income compared to its neighbors and suffers from a persistently high degree of inequality and poverty. The purpose of the current study is to contribute to a deeper understanding of growth volatility in Paraguay and to provide an input into the discussion on how to better manage it. In particular, the study will ask three questions: i) what are the sources of volatility in Paraguay? ii) How does growth volatility, in particular that arising from the strong dependence on the agricultural sector, impact the rest of the economy? iii) What are optimal policies for managing the types of volatility observed in Paraguay? This study’s quantitative analyses mainly rely on quarterly data available since the first quarter of 1994 (earliest available) and allow insights primarily into business cycle volatility. Wherever possible the study also shows a longer-term perspective based on yearly data. However, data restrictions do not allow for a rich analysis of these long-term volatility trends for Paraguay, and from a policy perspective business cycle volatility appears more relevant. The current study seeks to provide answers to the questions identified in three chapters: i) the first chapter covers the sources of volatility. It provides a description of stylized facts and an analysis of the sources based on structural vector autoregression (SVAR) analysis and a business cycle accounting exercise. ii) The second chapter addresses the effects of volatility with a particular focus on volatility arising from a strong dependence on the agricultural sector. This chapter is based on a qualitative analysis relying on 25 structured interviews with key players in the Paraguayan economy, as well as on a quantitative approach based on VAR analysis and a wavelet approach. iii) The third chapter presents policy options for managing volatility. In particular, it provides an overview and a discussion of cases of other Governments that have successfully managed volatility, similar to that observed in Paraguay. 3 See for example Loayza, Ranciere, Serven, and Ventura (2007), Athanasoulis and van Wincoop (2000), World Bank (2000) on the impact of volatility on welfare, Hnatkovska and Loayza (2005) and Calderon and Schmitt- Hebbel (2003) and Berument, Dincer, and Mustafaoglu (2011) on the growth impact, and Breen and Garcia Penalosa (2004), Garcia-Penalosa and Turnovsky (2004), or Huang, Fang and Miller (2012) for the impact on equality. 5 1.1. Stylized facts Paraguay’s GDP growth has been one of the most volatile in the region. The high volatility of GDP growth has coincided with a volatile macroeconomic environment. A large number of relevant economic variables have shown high levels of volatility in recent years, including the world interest rate, 4 Paraguay’s nominal exchange rate, its current account balance, public consumption and investment, credit to the private sector, agricultural GDP, rainfall and soy prices. While the agricultural sector was particularly affected, preceded by the increase in the volatility of soy prices and rainfall, non-agricultural GDP actually registered a decrease in volatility. Figure 1.1: Real GDP growth 14 12 10 8 6 Percent 4 2 0 -2 -4 -6 1951 1953 1955 1957 1959 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013* Source: Central Bank of Paraguay. Note:*Projection. Real growth in Paraguay has been more volatile than most other countries in Latin America and other regions in the past decade. One commonly used definition of volatility of economic growth is the standard deviation of real GDP growth rates or of the cyclical components of GDP. 5 According to these measures, Paraguay’s growth volatility was much lower than that of Latin American peer countries in the period from 1960 to 2000, whereas growth has been more volatile than most other countries in Latin America in the last decade. Real GDP growth varied by 4 percentage points in the period from 1960 to 2000, in the past decade it had a standard deviation of 5.5 (Table Annex 1.1). In contrast, many other countries in the region managed to reduce volatility, explaining the drop in the regional mean and median from 4.7 and 4.5 to 3.1 and 2.8 respectively. Paraguay’s MERCOSUR neighbor Brazil managed to reduce the variation of its real GDP from a standard deviation of 4.5 from the 1960 to 2000 period to 2.3 in the past decade, a pattern that the country shares with Bolivia, Chile, Colombia, Ecuador, Mexico and Peru. Also, the East Asia and Pacific region, Middle East and North 4 Measured by the U.S. 3-month treasury bill rate. 5 See for example Loayza, Ranciere, Serven, and Ventura (2007), Perry and Fiess (2006), Alouini and Hubert (2010), or Furceri and Karras (2007). 6 Africa, Sub-Saharan Africa, South Asia, and Paraguay’s peer group of lower middle income countries have all seen a reduction in growth volatility in the past decade (Figure 1.2). While countries in Europe and Central Asia, OECD members, and higher middle income countries share the trend of higher volatility in the period from 2000 to 2011, Paraguay’s growth volatil ity exceeds the ones in these regions. In fact, Venezuela, Argentina, and Uruguay are the only three countries in the period from 2000 to 2011 that have observed higher levels of growth volatility than that experienced by Paraguay. Figure 1.2: Volatility over time, Paraguay in regional comparison Panel a: Standard deviation (GDP growth)—1960-1999 Panel b: Standard deviation (GDP gap)— 1960-1999 Paraguay Paraguay LAC mean (excl. Paraguay) LAC mean (excl. Paraguay) LAC median (excl. Paraguay) LAC median (excl. Paraguay) East Asia & Pacific* East Asia & Pacific (all income… Europe & Central Asia* Europe & Central Asia* Middle East & North Africa* Middle East & North Africa* South Asia South Asia Sub-Saharan Africa* Sub-Saharan Africa* Lower middle income Lower middle income Upper middle income Upper middle income OECD members OECD members 0 1 2 3 4 5 6 0 1 2 3 4 5 6 Panel c: Standard deviation (GDP growth)—2000-2011 Panel d: Standard deviation (GDP gap)— 2000-2011 Paraguay Paraguay LAC mean (excl. Paraguay) LAC mean (excl. Paraguay) LAC median (excl. Paraguay) LAC median (excl. Paraguay) East Asia & Pacific* East Asia & Pacific (all income… Europe & Central Asia* Europe & Central Asia* Middle East & North Africa* Middle East & North Africa* South Asia South Asia Sub-Saharan Africa* Sub-Saharan Africa* Lower middle income Lower middle income Upper middle income Upper middle income OECD members OECD members 0 1 2 3 4 5 6 0 1 2 3 4 5 6 Source: World Development Indicators, and Central Bank of Paraguay, staff calculations. Note:*All income levels. While Paraguay has experienced clusters of high volatility in the past, fluctuations never quite reached the levels observed in recent years. In 2009, the economy contracted by 4 percent, the worst outcome in Paraguay’s recorded history, and rebounded to 13.1 percent in 2010, the best outcome ever observed. Shortly thereafter, Paraguay has been experiencing a similar pattern of extreme fluctuation, with a -1.2 contraction in 2012 and a projected recovery of 10.5 percent in 2013. This translates into extreme year-on-year (y-o-y) changes in terms of percentage points of growth: drops of 10 and 9 percentage points in 2009 and 2011 respectively, and increases of 17 and 14 percentage points in 2012 and 2013 respectively. The period in the past that comes closest to these extreme fluctuations is the end of the 1970s and the beginning of 7 the 1980s, which coincides with the construction and then completion of the Itaipú dam and hydro power plant (from 1975 to 1982). Growth dropped from 9.2 percent in 1981 to -1.4 in 1982. This period represented a transition between a period of high growth due to the impulse of the Brazilian Paraguayan construction project, approximately 4 times the size of Paraguay’s GDP at the time, and the period of low growth thereafter. Yet, GDP did not oscillate between sharp contractions versus fast expansions from year to year. Figure 1.3: GDP—breakpoints of volatility Figure 1.4: Agricultural GDP—breakpoints of quarterly y-o-y GDP growth using Inclan, of volatility of quarterly y-o-y GDP growth Tiao (1994) using Inclan, Tiao (1994) 8 25 Stddev. of agriculture GDP growth 12 7 36 20 St ddev. of GDP growth 6 26 7 Percent growth rate 5 15 16 Percent GDP growth 4 2 6 3 10 -4 2 -3 5 1 -14 -8 0 -24 0 Jan/ 95 Jan/ 96 Jan/ 97 Jan/ 98 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 Jan/ 95 Jan/ 96 Jan/ 97 Jan/ 98 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 GDP breaks stddev. Agriculture GDP breaks stddev. Source: World Bank calculations based on Central Source: World Bank calculations based on Central Bank, Paraguay. Bank, Paraguay. Three other past periods displayed significant y-o-y fluctuations, however not at the levels observed in the most recent past. First, between 1955 and 1961 GDP growth rates dropped twice by over 5 percentage points and also expanded twice by over 5 percentage points at the beginning of the Stroessner dictatorship (1954–1989) and after a period of high inflation (Cubas, Escobar, Franco, Olmedo, and Smith (2011)). Second, between 1966 and 1968, political uncertainties and some elements of democratization contributed to strong variations in economic growth. Between 1995 and 2002 Paraguay underwent a period of recurrent financial crises that went hand in hand with substantial fluctuations in growth. The biggest y-o-y change occurred between 1995 and 1996 when growth dropped from 6.8 percent in the first year to only 1.5 in the latter. A closer look at quarterly y-o-y real growth rates reveals that high volatility is a very recent phenomenon, with a significant increase in the fourth quarter of 2008 (Figure 1.3). The study relies on Inclan and Tiao (1994) to identify structural breaks in the volatility of the analyzed time series, a method that performs well with the type of data used.6 Before the break, the standard deviation of quarterly y-o-y growth since the first quarter of 1994 amounts to 4 percentage points, after the break it shifts up to 7 percentage points. 6 The Inclan-Tiao test is characterized by its simplicity and independence from estimated long-run variance, which make the test robust to time period selection, and it also performs well with shorter time series compared to other tests such as Kokoszka-Leipus (2000) or Quandt (1960) and Andrews (1993). 8 The dynamic of GDP growth mirrors an increase in the volatility of agricultural GDP in the fourth quarter of 2008, a sector whose weight has increased over time. The Inclan Tiao test applied to agricultural GDP growth reveals a break point in volatility also in the fourth quarter of 2008 (Figure 1.4). Before the fourth quarter of 2008, the standard deviation of real agricultural GDP growth was 6 percentage points, afterwards it shot up to 22. While the share of agriculture in total GDP amounted to about 12 percent in the second half of the 1990s it increased significantly to over 18 percent in 2010 and 2011 (Figure 1.5). Figure 1.5: Share of agriculture in GDP Figure 1.6: Growth volatility by economic sector 25 25 20-qurater standard deviation of quarterly y-o-y growth 20 20 Percent of GDP 15 15 10 10 5 5 0 0 2002:I 2005:I 2008:I 2011:I 2000:III 2003:III 2006:III 2009:III 2012:III 1999:IV 2001:II 2002:IV 2004:II 2005:IV 2007:II 2008:IV 2010:II 2011:IV 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Paraguay LAC Primary Secondary Tertiary Source: WDI. Source: BCP. An increase in rainfall volatility preceded the breakpoint of agricultural GDP volatility, this is relevant given that only 2 percent of agricultural surfaces are cultivated using irrigation. The instability of rainfall increased in 2007 and remained high until 2013 (Figure 1.7).7 Given the agricultural production cycle in Paraguay, rainfall from December the previous year and January of the current year are particularly relevant for the harvest. Figure 1.6 displays annual data aggregating rain data from those months. The correlation between the cyclical component of rainfall and agriculture GDP is 0.69, showing the high dependence of agriculture production on weather conditions (see Figure 1.8.). Despite the high and increasing correlation between the climate cycle and agricultural production, only 2 percent of the surface used for agricultural is irrigated.8 7 No formal volatility break test was applied because the length of the yearly time series does not allow for it. 8 United Nations Development Program (2006). 9 Figure 1. 7: Volatility of rainfall Figure 1. 8: Rainfall and agriculture GDP 1800 2500 15 1600 2000 10 1400 1500 1000 5 Percent of GDP 1200 Standard deviation Milimeters 1000 500 0 0 800 -500 -5 600 -1000 -10 400 -1500 200 -15 -2000 0 -2500 -20 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 1991 1993 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 Rain (cycle) Agriculture cycle (cycle as a share of GDP, RHS) Source: DINAC and BCP. Rain is considered for the months of December from the previous year and January of the same year because of their importance in the main agricultural soy product. All weather stations of the Eastern part of Paraguay are considered. Paraguay’s agricultural sector concentrates on a few products and export destinations. Soy and beef alone made up an average of 34 percent of total exports over the past 5 years; and exports to Brazil and Argentina alone reached almost 50 percent of total exports in the period since 2008 (Figure 1.9 and Table 1.1). Overall, agricultural GDP in Paraguay is much more volatile than the aggregate or than non-agricultural GDP. Quarterly y-o-y agricultural GDP growth has fluctuated by 12 percentage points since the first quarter of 1994, contrasting with overall GDP that showed a standard deviation of 4.7 percentage points, and with non-agricultural GDP which varied by 4 percentage points (Figure 1.5). In particular, the wedge between agricultural and non-agricultural GDP volatility has increased in recent years because non-agricultural GDP is one of the few macroeconomic aggregates that have not become more volatile but have remained relatively stable. The Inclan Tiao test did not identify any structural break in the volatility of this variable. Despite these facts the growth of agricultural GDP has exceeded the growth of aggregate GDP and of non-agricultural GDP. Quarterly y-o-y growth of agricultural GDP amounted to 4 percent compared to 2.7 percent of aggregate GDP and 2.5 percent of non-agricultural GDP (Table Annex 1.2). Paraguay’s macro-economic environment has become more volatile with shifts in the volatility of soy prices and world interest rates preceding those of other variables. Most macroeconomic variables underwent an increase in volatility during 2007 and 2008, some even increased before that (Table Annex 1.2). In light of the potential interaction between different economic aggregates, an interesting sequence is that soy prices became more volatile in the third quarter of 2003, and world real interest rates became more volatile in the fourth quarter of 2007, and that these increases preceded the increases of volatility in the nominal exchange rate in the first quarter of 2008; of public investment in the second quarter of 2008; of overall GDP and agricultural GDP in the fourth quarter of 2008; and of public consumption in the first quarter of 2009. Also the increase in the volatility of the current account balance in the first quarter of 2007 was preceded by an increase in volatility in soy prices. It is also important to note that there were additional increases, like that of inflation in the second quarter of 1995; public consumption in 10 the second quarter of 2000; or credit to the private sector in the fourth quarter of 2002. This illustrates the need to carefully assess causalities and to take into account relevant additional factors in an analytical assessment when searching for the sources of growth volatility in Paraguay. Such analysis is provided in the second section of this chapter. Figure 1. 9 Exports by product Table 1.1: Export by destination 100 50 Percent of total exports 80 40 60 30 Percent of GDP Total Continental Rest of the Argentina Brazil Uruguay MERCOSUR China World 40 20 average 13 47 6 66 1 33 since 2000 20 10 average 11 37 1 49 1 50 since 2008 0 0 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Other Grains Beef Soy beans Electricity Total in percent of GDP(RHS) Source: Central Bank, Paraguay. There are only a few exceptions to this increase in volatility; the main exception is tax revenues. Variables whose volatility has not changed over the period since the first quarter of 1994 include total and private investment, private consumption, oil prices, non-agricultural GDP, Paraguay’s real interest rate, the real exchange rate, total Government revenues, beef prices, and terms of trade. The only variable that displayed a decrease in volatility that was not followed by a later increase is tax revenues. The single breakpoint is the third quarter of 2004 coinciding with the 2004 tax policy and administration reform in Paraguay. While it is too early to tell whether high volatility in Paraguay is temporary or permanent, and there is no indication of a causal link explaining the increase in volatility, some structural changes are in line with the idea that volatility is there to stay. In particular, the increased weight of the agricultural sector and its concentration on a few products exposes the economy more to fluctuations in rainfall which seem to be increasing in the context of climatic change. It also renders the economy vulnerable to fluctuations in commodity prices. Yet, it is unclear whether commodity prices will remain as volatile as they have displayed clusters of heightened volatility in the past.9 1.2. Sources of volatility External shocks are the main source of growth volatility in Paraguay, with foreign demand explaining around 30 percent of GDP volatility, 20 percent is explained by the world interest rate, and 3 percent comes from terms of trade. Domestic variables also contribute to the volatility with 25 percent stemming from shocks to real GDP, 15 percent from shocks to 9 See Calvo-Gonzalez, Shankar, and Trezzi (2010). 11 investment, and 3 percent from pro-cyclical fiscal and monetary policy. Pro-cyclicality is defined as a positive response of government spending to an exogenous expansionary business cycle shock. In developing countries it is usually linked to limited access to credit in downturns, lax fiscal stances in good times, and burdensome bureaucratic processes. While Paraguay shares this pattern there is indication that the fiscal stance was counter-cyclical during the contractions in 2009 and 2012. In contrast to overall GDP volatility, fluctuations of agricultural GDP originate to a large extent from domestic shocks, with weather related shocks to agricultural output itself accounting for more than 50 percent. About half of Paraguay’s GDP growth volatility stems from external shocks while the other half originates in domestic shocks. These findings emerge from a variance decomposition based on a structural VAR analysis guided by economic theory for model specification. Hnatkovska and Koehler-Geib (2013) presents a detailed description of the model specification and a rational for variable selection. The variables included in the VAR are chosen to capture those factors identified in the literature as important determinants of business cycles in developing countries. External variables are: world interest rate (as measured by the US 3 month treasury bill rate), and foreign demand (as measured by the trade weighted GDP of Argentina, Brazil, Chile and Uruguay, the remaining export share is allocated to the US and its GDP).10 Domestic variables are: real GDP, total investment, Government consumption as measure of fiscal policy, domestic short-term interest rate as a measure of monetary policy, and the real trade balance to GDP ratio. The weights of the different variables reflect the variance decomposition at 12 quarters which is where the percentages for most cases stabilize and is the midpoint of the of the range between 6 and 32 quarters that researchers typically define as frequencies for business cycle movements (Baxter and King (1999)) (Table 1.2).11 Quarters TOT r_us Log(yf) Log(gc) Log(inv) Tby r Log(y) External Domestic 1 0.043 0.058 0.071 0.004 0.248 0.047 0.022 0.506 0.173 0.827 4 0.026 0.143 0.308 0.011 0.163 0.062 0.012 0.274 0.477 0.523 8 0.024 0.191 0.300 0.019 0.148 0.058 0.011 0.248 0.515 0.485 12 0.025 0.195 0.299 0.020 0.147 0.058 0.011 0.247 0.518 0.482 16 0.025 0.195 0.299 0.020 0.147 0.058 0.011 0.247 0.518 0.482 20 0.025 0.195 0.299 0.020 0.147 0.058 0.011 0.247 0.518 0.482 Source: Hnatkovska and Koehler-Geib (2013) 10 Kose, Otrok, and Prasad (2012) and Kose, Otrok, and Whiteman (2008) find a stronger role of domestic factors. This difference appears to mainly stem from a sample period that only goes up until 2005 only. Raddatz (2007) also finds a predominant role of external variables in a paper that examines whether the differences in output volatility between Latin America and other regions result from volatility of external shocks or from a more pronounced response to these. Podpiera and Tulin (2012), focusing on the role of financial external variables find a relevant role of external factors. 11 Table 1.2 presents the results of the variance decomposition based on the estimated SVAR system at different horizons. The variance decomposition allows quantifying the contribution of each shock to the variance of forecasting error for output. 12 The most important external factor, foreign demand, explains around 30 percent of Paraguayan growth volatility, and terms of trade another 3 percent, a fact that may be related to Paraguay’s agricultural sector. The significance of these shocks may arise from the sector’s weight in the economy and its concentration on a few export products (see Section 1.1). As argued in Hnatkovska and Koehler-Geib (2013) concentration in terms of products and export destinations renders countries more vulnerable to terms of trade and foreign demand shocks. Figure 1.10: Correlation between trade balance Figure 1.11: Impulse response of trade balance and world interest rate to a shock to the world interest rate .15 .06 r_us shock .004 .04 .1 Reat int. rate, world .002 .02 TB/GDP .05 0 0 0 -.02 -.002 -.04 -.05 0 5 10 15 20 quarters 1995q1 2000q1 2005q1 2010q1 95% CI IRF TB/GDP Reat int. rate, world Graphs by irfname, impulse variable, and response variable Source: Hnatkovska and Koehler-Geib (2013) The second most important external factor, the world interest rate, explains around 20 percent of Paraguayan growth volatility through its impact on portfolio reallocation, on commodity prices, the economic condition of trade partners, and remittances. At a first glance it may appear surprising that GDP growth of a country which is not heavily represented in international capital markets varies this much with the international interest rate. Yet, there are mainly four channels that explain the link. First, when international interest rates go up, foreign investors may shift out of Paraguayan assets inducing a contraction or even a reversal of capital inflows. Calvo, Leidermann, and Reinhart (1993) and Gavin, Hausmann, and Leidermann (1995) show these “pull� effects for emerging economies. The positive correlation between Paraguay’s trade balance and the world interest rate of 0.2, as well as the significant positive effect in the impulse response of the trade balance to an interest rate shock controlling for other shocks, suggest that this channel is relevant for Paraguay (Figures 1.10, 1.11 and Table 1.3).12 Second, the world interest rate variable may pick up some of the effects of commodity price changes on the Paraguayan economy despite controlling for terms of trade in the SVAR analysis. 13 One reason for the link between world real interest rates and commodity prices is that higher interest rates reduce the speculative demand for commodities inducing lower prices (see for example 12 The trade balance functions as a proxy for net international financial flows. The two are closely related through the balance of payments identity as trade has to be balanced every period in the absence of international financial flows. 13 The recursive identification scheme does not allow for a contemporaneous correlation between the two variables. Note that changing the order of the two variables in the SVAR will not resolve the simultaneity problem. In fact, the results remain robust to a change in the ordering of the two variables. 13 Frankel (2008)).14 Figure 1.12 shows the negative relationship between the international interest rate and the TOT for Paraguay. Indeed, the correlation is equal to -0.44 during the sample period (Table 1.3). Third, a higher world interest rate may also affect the Paraguayan economy indirectly by weakening the economic conditions of its major trade partners – Brazil, Argentina, Chile, Uruguay – who are significantly exposed to the world financial markets through sovereign borrowing. The foreign demand variable used in the SVAR analysis may not fully capture demand shocks from these countries if nominal rigidities exist in the goods or factor markets, if a significant informal sector exists, or if supply and demand shocks are correlated. Fourth, a higher world interest rate, may lead to lower inflows of remittances to Paraguayan households from abroad through its impact on trading partners. While external shocks like those to commodity prices and world real interest have hit all countries, their output response reflects country specific conditions. In particular, the strong dependence on a few goods and services, a narrow tax base, and economic policies seem to play a role in the propagation of external shocks and their output response in Paraguay. Figure 1.12: Correlation between TOT and world interest rate .06 200 .04 Real int. rate, world .02 150 0 -.02 100 -.04 1995q1 2000q1 2005q1 2010q1 Real int. rate, world Terms of trade (Pex/Pim) Source: Hnatkovska and Koehler-Geib (2013) GDP GDP non- Int Int GDP Gov Gov GDP agri agri Inv rate rate US foreign TOT cons inv GDP 1 GDP agri 0.39 1 GDP non-agri 0.97 0.17 1 Inv 0.89 0.33 0.87 1 Int rate -0.17 0.00 -0.18 -0.23 1 Int rate US 0.13 -0.15 0.18 0.00 0.41 1 GDP foreign 0.82 0.30 0.80 0.74 -0.23 0.24 1 TOT -0.56 0.08 -0.63 -0.35 0.25 -0.44 -0.74 1 Gov cons 0.85 0.01 0.91 0.86 -0.09 0.09 0.65 -0.42 1 Gov inv 0.10 -0.18 0.16 0.02 0.14 0.02 -0.03 0.07 0.26 1 Source: Hnatkovska and Koehler-Geib (2013) 14 A large economic literature analyzes the theoretical and empirical link between world interest rates and interest (see for example Calvo (2008), Ratnovski and Mihet (2012), Byrne, Fazio, and Fiess (2012), and Frankel and Rose (2010)). 14 In terms of domestic variables, 25 percent of GDP volatility stems from shocks to real GDP, 15 percent from shocks to investment, and 3 percent from pro-cyclical monetary and fiscal policies. Investment in Paraguay is one of the most volatile domestic variables (Table Annex 1.2) and is highly pro-cyclical with a correlation between the cyclical components of investment and GDP amounting to 0.9 (Table 1.3). Monetary policy is captured by the short-term real interest rate and explains 1 percent of GDP volatility. Recent work shows that real interest rates tend to be counter-cyclical in developing countries, while they tend to be pro-cyclical in developed economies (see, for instance, Neumeyer and Perri (2005), Uribe and Yue (2005)). A prominent explanation for pro-cyclicality includes distortions in factor markets: for example, firms may have to pay for part of the factors of production before production takes place, creating a need for working capital. This is also the case for Paraguay, where GDP and the real interest rate are negatively correlated, with an unconditional correlation equal to -0.2. This correlation, however, is somewhat smaller than the corresponding number in the other Latin American countries: it is equal to -0.63 in Argentina; -0.49 in Mexico; and -0.38 in Brazil (see Neumeyer and Perri (2005)). The fiscal policy stance is captured by Government consumption and investment which could be used as tools for counter-cyclical policy. Yet data shows that they have not been counter-cyclical in Paraguay. From 1994 to 2011 the correlation between the cyclical components of government consumption and GDP is 0.9; this explains 2 percent of GDP volatility in Paraguay. Government investment is less pro-cyclical and has a correlation with GDP of 0.1. Figure 1. 13: Business cycle fluctuations in Figure 1.14: Business cycle fluctuations in Paraguay—Government consumption versus Paraguay—Government investment versus GDP GDP .3 .1 .1 .5 .2 .05 .05 Gov Cons (log) 0 Gov Inv (log) .1 GDP (log) GDP (log) 0 0 0 -.5 -.05 -.05 -.1 -.2 -.1 -.1 -1 1995q1 2000q1 2005q1 2010q1 1995q1 2000q1 2005q1 2010q1 GDP (log) Gov Cons (log) GDP (log) Gov Inv (log) Source: Hnatkovska and Koehler-Geib (2013) Pro-cyclicality of fiscal policy in developing countries is usually linked to limited access to credit in downturns, lax fiscal stances in good times, and burdensome bureaucratic processes. Pro-cyclicality is defined as a positive response of Government spending to an exogenous expansionary business cycle shock. Gavin and Perotti (1997) showed that this is the case in Latin America. Talvi and Végh (2005) then claimed that pro-cyclical fiscal policy is not only a Latin American phenomenon, it is present in the entire developing world. In a recent study, Ilzetzki and Végh (2008) revisit the evidence using a sample of 49 countries while allowing for a reverse causality running from fiscal policy to GDP. They show that fiscal policy is indeed pro-cyclical in developing countries. One reason for this pro-cyclicality could be frictions in international credit markets that prevent developing countries from borrowing in bad 15 times ((Gavin and Perotti (1997), Caballero and Krishnamurthy (2004), Mendoza and Oviedo (2006), and others); another reason originates from a political economy perspective, and proposes that good times encourage fiscal profligacy ((Tornell and Lane (1998), Talvi and Végh (2005), and others); the third reason rests in delays in the implementation and execution of fiscal policies in developing economies. While fiscal policies over the last two decades were pro-cyclical in Paraguay, a look at the data suggests that they were counter-cyclical for a short period of time during the contractions of 2009 and 2012. Public sector demand expanded when private demand, and as a consequence economic growth, collapsed in all four quarters of 2009 (see Figure 1.15). The decomposition of real growth into the components of aggregate demand reveals that public demand components together contributed positively to real growth in the four quarters of 2009 while private demand contracted heavily (see Figure 1.15).15 The expansion of public demand was based on strong increases in both public investment and consumption (see Figures 1.13 and 1.14). The reason for policies being counter-cyclical during the 2009 crisis was that Paraguay had built up buffers through prudent fiscal policies in prior years and had access to financing, from international institutions. Paraguay shares this pattern of a recent move towards counter- cyclical fiscal policy with other developing countries. Vegh and Vuletin (2013) document that around one third of developing countries were able to conduct countercyclical fiscal policy over the last decade. However, public demand ceased to be counter-cyclical by the first quarter of 2010, when it expanded at the same time as private demand was already recovering strongly. Only in the third quarter of 2010 did public demand contribute negatively amidst a rapid private sector expansion. A similar pattern can be observed in 2012 when there was the same challenge of withdrawing expansionary expenditure fast enough as private sector growth was recovering. Figure 1.15:Contribution of public and private demand components to real GDP growth 20 Percent y-o-y real growth and 15 percentage contribution 10 5 0 -5 -10 -15 2003 Q1 2007 Q3 2012 Q1 2001 Q1 2001 Q3 2002 Q1 2002 Q3 2003 Q3 2004 Q1 2004 Q3 2006 Q1 2006 Q3 2007 Q1 2008 Q1 2008 Q3 2009 Q1 2009 Q3 2010 Q1 2010 Q3 2011 Q1 2011 Q3 2012 Q3 2005Q1 2005Q3 Public demand Private Demand GDP growth Source: Central Bank of Paraguay. In terms of sectors, variations in agricultural GDP contribute one quarter to overall GDP volatility; three quarters are explained by non-agricultural GDP. With the help of variance 15 Public demand components comprise public consumption, public investment, and the share of the public sector in imports and changes in inventories. 16 decomposition Hnatkovska and Koehler-Geib (2013) find that a 1 percent increase in aggregate GDP is accompanied by a 0.25 percent increase in agricultural GDP and 0.75 percent increase in non-agricultural GDP. This finding is consistent with the high volatility of agricultural GDP in recent years and the fact that agriculture has a share of about 20 percent in overall GDP. So while more volatile, the contribution of agricultural GDP remains limited by its size in aggregate GDP. Table 1.4: Variance Decomposition of Agricultural and Non-Agricultural GDP volatility Agriculture Log Log Log Log Log quarters TOT r_us (yf) (gc) (inv) Tby r (yA) (yNA) Ext. Dom. 1 0.047 0.010 0.031 0.053 0.031 0.000 0.009 0.819 0.000 0.088 0.912 4 0.038 0.023 0.208 0.149 0.042 0.015 0.010 0.513 0.001 0.270 0.730 8 0.038 0.026 0.228 0.162 0.040 0.015 0.010 0.479 0.002 0.292 0.708 12 0.038 0.026 0.229 0.162 0.040 0.015 0.011 0.477 0.002 0.293 0.707 16 0.038 0.027 0.229 0.162 0.040 0.015 0.011 0.477 0.002 0.293 0.707 20 0.038 0.027 0.228 0.162 0.040 0.015 0.011 0.477 0.002 0.293 0.707 Non-Agriculture Log Log Log Log Log quarters TOT r_us (yf) (gc) (inv) Tby r (yA) (yNA) Ext. Dom. 1 0.185 0.037 0.036 0.014 0.191 0.059 0.061 0.006 0.412 0.258 0.742 4 0.134 0.151 0.170 0.028 0.136 0.070 0.033 0.008 0.270 0.454 0.546 8 0.120 0.221 0.159 0.027 0.124 0.068 0.030 0.010 0.243 0.500 0.500 12 0.120 0.227 0.157 0.027 0.122 0.067 0.030 0.010 0.241 0.504 0.496 16 0.120 0.228 0.157 0.027 0.122 0.067 0.030 0.010 0.240 0.505 0.495 20 0.120 0.228 0.157 0.027 0.122 0.067 0.030 0.010 0.240 0.505 0.495 Source: Hnatkovska and Koehler-Geib (2013). In contrast to overall GDP volatility, fluctuations of agricultural GDP originate mainly from domestic shocks, with weather related shocks to agricultural output itself accounting for almost half. A variance decomposition based on an SVAR specification that includes agricultural and non-agricultural GDP instead of aggregate GDP in Hnatkovska and Koehler- Geib (2013) shows that 70 percent of the volatility of agricultural GDP can be attributed to domestic factors, while the rest is explained by external factors. The most important domestic variable is agricultural output itself; shocks to this variable account for 47 percent of volatility in agricultural GDP (Table 1.4). While Hnatkovska and Koehler-Geib (2013) provide evidence of a correlation of 0.36 between structural shocks to agricultural output and rainfall, a complementary analysis by Berument (2013) goes more into detail and finds that agricultural output significantly responds to shocks to rainfall. Moreover, the analysis shows the significant impacts of international soy and beef prices. This is not surprising given that Paraguay is a price taker in international commodity markets. 16 Soy prices even impact non-agricultural GDP, an effect which could run through the indirect impact of disposable income or the inclusion of soy bean processing into the value chain. Machinery inputs were found not to be relevant. There is no data on land prices in Paraguay, and therefore the impact of land prices on agricultural output could 16 Section 2.1 provides a detailed description based on Berument (2013). 17 not be evaluated. Berument (2013) provides a detailed description of the underlying VAR analysis of weather and price impacts on agricultural GDP and non-agricultural GDP. Shocks to other domestic variables such as investment, Government consumption and real interest, show similar effects to those for aggregate GDP, except that their impacts on agricultural GDP are larger. For instance, a 1 percent shock to Government consumption leads to a 1 percent decline in agricultural GDP. This confirms the idea that pro-cyclical fiscal policy may have detrimental effects on domestic conditions, especially in agriculture. Also in the case of the external variables such as terms of trade, world interest rate, and foreign output, the qualitative effects on agricultural GDP are the same as those on aggregate GDP. Yet quantitatively, the effects of terms of trade and world interest rate on agricultural GDP were significantly larger. Impulse responses show for instance that a positive shock to the terms of trade leads to a 1 percent increase in real agricultural GDP, while it leads to about 0.2 percent increase in aggregate real GDP. Similarly, unanticipated shocks to the world interest rate lead to a reduction in Paraguayan agricultural GDP and the effects of these shocks are significant and larger than on aggregate GDP. The shocks to foreign demand, in contrast, have a somewhat smaller effect on agricultural GDP than on aggregate GDP. 1.3. The role of factor markets Despite a significant decrease, remaining rigidities in factor markets and limited mobility across sectors reduce the economy’s capacity to buffer shocks and exacerbate business cycle fluctuations and hence volatility. These findings stem from a model-based examination of the sources of business cycle volatility in Paraguay covering the period from 1991 to 2010 (Hnatkovska and Koehler-Geib (2013)). More precisely this is a business cycle accounting analysis based on the methodology of Chari, Kehoe, and McGrattan (2007) introducing time- varying wedges into a standard neoclassical growth model. The wedges represent frictions and distortions in labor and capital markets, and shocks to efficiency, government spending and trade balance. The model is calibrated for the Paraguayan economy to quantify the frictions and evaluate their contribution to GDP volatility. While labor market distortions have declined, firms’ access to credit have improved, and agricultural efficiency has increased, important challenges remain which suggest that shocks hitting Paraguay may rather be exacerbated than buffered. In particular, labor and capital returns between agriculture and non-agriculture remain large, suggesting limited factor mobility across sectors and the efficiency in the non-agricultural sector has shown no signs of improvement, to the contrary has been deteriorating. Labor market distortions have become less important in Paraguay over the analyzed period. As argued in Hnatkovska and Koehler-Geib (2013), labor market frictions may arise from payroll taxes, distortions due to unionization, collective bargaining, hiring and firing costs, or sticky wages. An analysis of the Doing Business survey available back until 2006 suggests that the indicator of hiring and firing costs has remained stable over time. At the same time, minimum apprentice wages have increased substantially in Paraguay during this time. This would be consistent with the idea that more young workers could be attracted into work force participation thus reducing frictions in the market. 18 Firms’ access to credit has improved over time while financing constraints facing households remain pronounced and time-varying. Financing constraints affecting firms’ investment decisions have decreased notably with significant volatility around the trend. One way to evaluate this finding would be to look at the dynamics of private credit to businesses and households. The available data on aggregate credit to the private sector as percent of GDP shows an expansion of 35 percent between 1994 and 2011, suggesting an improvement in credit market conditions in Paraguay during this period. Moreover the Doing Business survey also provides evidence of improved credit conditions in Paraguay during the period between 2004 and 2012 (see Hnatkovska and Koehler-Geib (2013) for more details). In terms of sectors, the efficiency of agriculture has been continuously increasing, while efficiency of non-agriculture has been decreasing. These trends reflect the fact that measured agricultural productivity has been improving during the period from 1991 to 2010, averaging 3 percent per year; while the measured non-agricultural productivity has been falling, averaging - 1.5 percent annually. This productivity measure does not only reflect total factor productivity but also include human capital, weather conditions, omitted inputs, misallocation of resources, institutional factors, and in fact everything that may lead to inefficient human and physical capital stocks in each sector. Significant distortions remain in terms of returns of labor and capital between the agricultural and the non-agricultural sectors, suggesting that factor mobility remains limited in Paraguay, preventing the equalization of value marginal products across sectors. The analysis shows a 5-fold relative gap in favor of non-agricultural labor returns. While a small improvement after 2005, suggests some recent improvements in workers’ returns in agriculture, the gap relative to non-agricultural workers remains significant. The gap in sectoral returns to capital shows more variability, but returns to capital remained in favor of the non-agricultural sector. These remaining frictions reduce efficiency of the Paraguayan economy and prevent its capacity to buffer shocks that hit the economy. In particular, the low factor mobility between sectors may lead to shocks being exacerbated rather than buffered. 19 Chapter 2: The effects of growth volatility in Paraguay with a focus on volatility originating in the agricultural sector Growth volatility impacts Paraguay’s economy in various ways, it impacts i) the agricultural sector; ii) other sectors; and iii) macroeconomic aggregates such as investment; tax revenues; and poverty and equity. The main sources of volatility in the agricultural sector can be categorized into shocks to production and shocks to markets. Shocks to production include variations in rainfall, investment levels, and disease outbreaks (e.g., foot and mouth disease). Shocks to markets include commodity price variations, the closing of markets in the case of disease outbreaks, and fluctuations in the prices of imported inputs like fertilizers and pesticides. Overall, market participants in Paraguay report that a lack of information and knowledge with regard to the patterns and impacts of volatility on the economy is the biggest challenge to operating within this environment. Within the agricultural sector the sources of volatility manifest themselves through shocks that impact the level of infrastructure and R&D investment; cause payment delays; and trigger the use of diversification strategies. In terms of other sectors, the volatility of agricultural GDP mainly affects those economic activities that provide inputs such as machinery or storage and transport services, but it also affects the financial services and construction sectors. While the effect of agriculture on the services and construction sectors are small, they are statistically significant. No significant effects are found running from agriculture to mining and industry or to the electricity and water sectors. In terms of macroeconomic aggregates, the exchange rate and levels of employment fluctuate as a consequence of shifts in agricultural exports. There is some indication that private consumption plays a role in propagating the impact of agricultural GDP through the economy, impacting non-agricultural GDP. Investment levels are lower, fiscal revenues are indirectly affected, and the reduction of poverty and inequity are generally slower than in countries with lower levels of volatility. The analysis of this chapter relies both on quantitative methodologies in the form of vector auto regression (VAR) analysis and structured interviews with 25 key players in the Paraguayan economy both from the agricultural and non-agricultural sectors. The findings of the structural interviews are in line with the econometric findings. Yet, based on the limited sample size of the interviews, these findings are considered as supporting evidence rather than stringent proof of the hypotheses in the analysis. As this study focuses mainly on business cycle volatility it relies on quarterly data which is available from the first quarter of 1994 until the fourth quarter of 2011.17 Section 2.1 will present the propagation of shocks within the agricultural sector, section 2.2 will address the impact on other sectors, and section 2.3 will describe those effects of volatility on macroeconomic aggregates that originate in the agricultural sector. 2.1 Propagation of shocks within agriculture 17 See Berument (2013), Berument (2013a) and Hnatkovska and Koehler-Geib (2013) for a detailed description of the quantitative analysis. 20 Production and market shocks associated with climate, levels of investment, sanitary conditions in the livestock sector, and commodity prices cause volatility in agricultural GDP. Key actors in the agricultural and non-agricultural sectors are very conscious of these sources of volatility, which are consistent with the findings of the SVAR analysis presented in section 1.2.18 Interviewees for the qualitative study were selected to represent main economic activities and business occupations. They were grouped into the following categories: i) corporate farming; ii) companies along the agricultural value chain including production, provision of inputs such as machinery, and storage; iii) transport; iv) financial services including banking and insurance; iv) utilities; v) import and export; and vi) think tanks. How do the shocks identified affect the agricultural sector and how do they percolate through the economy? Climate impacts agriculture through its effect on productivity and on fluvial transport conditions. Rainfall and soil temperature create volatility in agricultural production through their impact on productivity per hectare.19 Rainfall also has an important impact transport through its impact on the navigability of the Paraguay and Paraná rivers, which are Paraguay’s main means of transport for bringing agricultural exports to international markets. Wavelet analysis confirms the impact of rainfall on agricultural output. Figure 2.1: Wavelet analysis of rainfall and agricultural GDP 0.8 3 4 0.6 6 Period 0.4 8 11 0.2 16 10 20 30 40 50 60 Time Source: Berument (2013a) Rainfall leads agricultural GDP by two to three quarters and by seven to eight quarters (Figure 2.1). These results stem from a wavelet analysis relating the level of quarterly rainfall to agricultural GDP. Wavelet analysis resembles Fourier type analysis of time series which involves decomposing a time series into an array of sinusoidal waves and checking for the linkages between the series of interest at similar wavelength. This allows pinpointing at what frequencies the series of interest move together, or at what frequencies one time series leads the other. In contrast to Fourier type analysis, Wavelet analysis does not have an indefinite number 18 In the SVAR analysis presented in section 1.2, commodity prices are reflected in the terms of trade, private investment is included directly, and climate, innovation, and sanitary conditions are subsumed as shocks to agricultural output itself. 19 Due to lack of data on soil temperature this variable could not be included in the quantitative analysis. 21 of sinusoidal waves, whereas, a time series is expressed in terms of wavelets (small waves) which have short durations.20 The heat map above summarizes the estimates of the relationship between rainfall and agricultural GDP. The vertical axis shows the period length, for instance a period length of 4 corresponds to oscillations over 4-quarter periods and a period length of 16 corresponds to oscillations over 16-quarter periods. The horizontal axis shows time, which is running from 1997Q1 to 2011Q4 in the subsequent analyses. The shift of colors from blue to red indicates a strengthening relationship and an upward directed arrow indicates the second variable leads the first one at given wavelength. The bold black contours (obtained using Monte Carlo simulations) indicate a significant relationship. The conic envelope can be viewed as the region where estimates have higher reliability. The map shows that rainfall cycles lead agricultural output in the short run with a 2- to 3-quarter wavelength indicating that rainfall boosts the volume of agricultural output within a year. The relationship that shows at 7- to 8- quarter wavelength could have to do with the pattern that one year of high productivity is often followed by a poorer performance. It could also be related to accumulation of water in the soil. Secondly, volatile and relatively low levels of investment, particularly investment on technological innovation, increase the volatility of agricultural GDP, or at least do nothing to reduce it. As pointed out in sections 1.1 and 1.2 investment is the most volatile domestic variable in Paraguay. The reasons behind the volatile investment environment may have to do with uncertainty about future growth in a volatile environment as will be explained later in this section. The volatility of investment and outcomes may therefore be a self-reinforcing cycle. Moreover, levels of investment remain low, and therefore opportunities for reducing volatility are missed. Genetic innovation, combined with bio-technological procedures, could reduce reliance on the climate if crops were more resilient and if crop rotation techniques were used. Limited investment in irrigation has a similar effect. Third, sanitary measures in the livestock sector are an important precondition for exports in this sector and failures in the safety measures create large fluctuations. Failure to implement acceptable measures of hygiene and the outbreak of foot and mouth disease in 2011, led to a drop in production and an exclusion of Paraguayan beef from Chile. As a consequence exports dropped significantly. Fourth, the price of soy and beef render agricultural GDP more volatile as Paraguay is a price taker in international markets and as farmers adjust their supply to expected prices. Impulse response functions based on a simple VAR relating Paraguay’s agricultural GDP to world agricultural raw material prices illustrate that prices significantly impact Paraguay’s agricultural GDP, while no significant feedback from Paraguayan agricultural GDP to world agricultural raw material prices can be detected (Figure 2.2). This analysis is based on annual data from 1964 to 2011. Similar relationships can be identified for the prices of soy and beef with agricultural GDP in quarterly frequency from 1994 to 2011.21 Moreover, wavelet analysis also confirms Paraguay’s price taker role in international commodity markets. 22 Increased 20 Berument (2013) provides more details on the methodology. 21 Berument (2013) provides details on the analysis. 22 See Berument (2013a). 22 volatility in commodities in recent years, has also affected agricultural GDP in Paraguay. Another link between commodity prices and agricultural GDP is explained by farmers’ supply responses to price expectations. Agricultural producers expand or reduce production if they expect high or low commodity prices. The supply elasticity of soy and beef production in Paraguay vis-à-vis respective commodities is relatively high, suggesting a fast and significant response (Favaro, Koehler-Geib, Picarelli, and Indaco (2013)). Figure 2. 2: Impulse response functions linking Paraguay’s agricultural GDP to world agricultural raw material prices23 Source: Berument (2013) Exchange rate fluctuations and inefficiencies in the market structure were also identified as sources of volatility, albeit of lesser concern. Beef producers pointed out that the strong seasonality of grain production induces strong variations in the exchange rate, which imposes financial uncertainties on producers of other products, including beef. The reason for this uncertainty is that a strong grain harvest leads to increased inflows in dollars and hence an appreciation of the Guaraní. This affects the beef value chain that also operates across currencies. While reefer companies pay beef producers in Guaranís, export prices are fixed in dollars. When the Guaraní appreciates, beef producers are negatively affected by the exchange rate change and in some cases are not able to fulfill commitments in Guaranís . In addition, small- and medium- sized farmers point out that the monopoly position of multinational grain producers introduces volatility, because these companies operate as price setters throughout the value chain, from the provision of inputs such as grains, fertilizers or pesticides, through the financing of production, and including transportation and storage. Market participants in Paraguay report a lack of information and knowledge of the patterns and impacts of volatility on the economy as the biggest challenge to operating in this environment. This creates uncertainty, which in turn leads to severe disincentives to 23 In the figure DLCMARM stands for the annual growth rate of world agricultural raw material prices and YAGRLCUG stands for the annual growth rate of Paraguay’s agricultural GDP. 23 investment. One important area where existing data is not analyzed in a systematic way is the weather. In Paraguay, different institutions with a variety of objectives collect data on the weather. DINAC, Dirección Nacional de Aeronáutica Civil, has the oldest and most complete database that has been collected with the intention of monitoring weather changes for aviation. The two bi-national hydroelectric power plants, Itaipú and Yacyreta, also monitor weather data to better control the water level in the reservoirs. Several agricultural associations such as FECOPROD, Federación de Cooperativas de Producción, collect climate data for agricultural purposes. Furthermore, South American weather and climate are also carefully watched by institutions with a global scope, for example the US National Weather Service, and the Climate Prediction Service. While all these data sources exist they are not analyzed and used in a consistent and coordinated way. This weakens the ability to adequately predict weather patterns for the purposes of agricultural production and therefore impacts on the ability to forecast GDP. Farmers and agricultural corporations reduce investment in infrastructure, innovation, and machinery as a reaction to uncertainty and to negative shocks. Volatility renders planning more challenging. Investment plans with fixed costs become obsolete during the production cycle, sometimes forcing a disinvestment or a significant financial loss at the end of the production cycle. Decisions based on a fall in production and exports affect subsequent agricultural cycles in most cases. Livestock firms, for example, reduce investment in genetic material or pasture improvement. Grain producers diminish investment in machinery, storage capacity, or expansion of farm land. Faced with a negative shock, payment delays as well as credit refinancing and restructuring are more commonplace. Farmers and agricultural corporations find it difficult to service financial commitments if harvests fall short of expectations. This results in payment delays, credit refinancing and restructuring. Overall, interviewees conjecture that volatility is at the root of high credit rates. Family and corporate farms as well as cooperatives react to the volatile environment by diversifying production. An important feature of the Paraguayan agricultural sector is that most land is cultivated by large corporations. However, a significant sector of family, small corporate farms, and cooperatives coexists with large firms. These smaller farms are engaging into diversification strategies to reduce the dependence on a single or only a few commodities. Namely, production is diversified into dairy products, small animal life stock, fruit and vegetables. This raises questions of the trade-off between risk diversification and scale of production. 2.2 The impact of volatility originating in the agricultural sector on other sectors Fluctuations in the agricultural sector impact the service and construction sectors in a statistically significant manner; the measurable impact is relatively small however. 24 Both, 24 Acosta-Ormaechea (2011) finds very little spill-overs from agriculture to other sectors in a VAR analysis that spans a very short time period from 2003 to 2010. 24 the quantitative VAR analysis by Berument (2013), as well as the qualitative analysis with structured interviews by Borda, Anichini, and Ramirez (2013), identify the service and construction sectors as those most affected. The VAR analysis uses quarterly data from the third quarter of 1994 to the fourth quarter of 2011. The specification of this simple VAR is to include rain, agricultural GDP, and the respective sectoral GDP. Two definitions of agricultural GDP are used. When a broad definition of agriculture is used in the VAR analysis, i.e., including cattle, fishery, and forestry, the impact on services and construction is positive and significant (Figures 2.3 and 2.4). When cattle, fishery, and forestry are excluded the impact on construction becomes insignificant. The relationship with services remains unchanged. Quantitatively, the impact is small however. In case of the narrow definition of agriculture a 12 percent expansion of agricultural activity induces a one percent expansion of the services sector. Irrespective of the concept of agriculture, no significant relationships could be established between agricultural GDP and mining and industry, or electricity and water (Berument 2013). Unconditional correlations as presented in Annex 1.4 also confirm these sectoral interrelations with a further breakdown of services by subsector. Together, the mining and industry, and electricity and water sectors represent around 23 percent of GDP. The small effect of agricultural GDP on the services and construction sectors as well as the fact that there is no significant effect from agriculture on an important part of the economy supports the fact that agricultural GDP has become much more volatile at the same time as the volatility of non-agricultural GDP has remained fairly stable. Figure 2.3: Impulse response functions linking Paraguay’s agricultural GDP to the construction sector25 25 In the figure RAIN stands for the quarterly growth rate of rainfall, AGR2G is the quarterly growth rate of agricultural GDP (broadly defined including cattle, fishery, and forestry), and JCONSA_log_d1 stands for the quarterly growth rate of the value added of the construction sector; wherever necessary seasonally adjusted series were used. 25 Source: Berument (2013) Companies that provide inputs to agriculture, for example, machinery or; veterinary products including genetic material; and seeds, tend to suffer late payments, demand shifts, and reduced capacity usage when agricultural production falls. Companies that provide machinery experience delays in payment and so provide help with refinancing. Veterinary companies experience shifts in demand towards lower quality products, for example cattle farmers switch to less expensive breeds, and the seed sector experiences reduced physical capacity usage, at times down to 40 percent of normal usage. Agricultural volatility may have contributed to the relatively low level of installed, static storage capacity in Paraguay. Storage services are divided into two categories: i) static storage in the form of large silos, and ii) non-static storage in form of small silos (e.g., silo bags). In Paraguay, CAPECO estimates that the static capacity of silos for soy in 2011 was close 6 million tons while production was significantly higher at 8 million tons. Storage capacity constraint may explain why soy beans are normally exported within a month after harvesting. Static storage requires medium- to long-term investment and high volatility introduces disincentives to do so. Providers of non-static storage see the demand for their products adjust as agricultural output changes. Figure 2. 4: Impulse response functions linking Paraguay’s agricultural GDP to the services sector26 Source: Berument (2013) 26 In the figure RAIN stands for the quarterly growth rate of rainfall, AGR2G is the quarterly growth rate of agricultural GDP (broadly defined including cattle, fishery, and forestry), and JSERSA_log_d1 stands for the quarterly growth rate of the value added of the construction sector. 26 The transport sector is impacted by strong variations in demand and as a result also by profit margins. Faced with falling demand for transport volumes, the share of fixed costs in total costs increases and profit margins drop. When agricultural production falls, shipping companies first notice a reduction in the amount of fuel transported. Subsequent stages see lower shipment volumes of agricultural output. Transport companies that belong to multinational corporations are particularly hard hit by a drop in production because they tend to specialize in the transport of a select group of commodities. The multinational corporations to which Paraguayan transport companies belong absorb the losses in the transport sector with earnings from other lines of business. The insurance sector in Paraguay experiences a strong bunching of risks, in particular during seasons where bad outcomes are expected; this leads to high policy rates. The insurance sector in Paraguay only offers a small range of products for the most important commodities of the economy. Rice production is not offered in insurance policies, but insurers are currently assessing its profitability with a view to future inclusion. A strong bunching of risks occurs related to producers’ expectations of their harvest outcomes. Producers typically don’t purchase insurance if they expect a good harvest, and only do so when expectations are grim. The result of this unpredictability in insurance coverage means that policy rates are high. Following a bad year, when there may even have been a natural disaster and the insurance companies have a high pay-out they raise insurance policy rates for the next year. Many producers decide against insurance and absorb the risk themselves, as they can compensate for one bad harvest with a good harvest the following year; yet if a bad cycle were to last for two years, the effects on producers and the economy would be severe. Interviews with 25 key players in the Paraguayan economy reveal that many producers in Paraguay do not apply any measures to mitigate agricultural volatility risks. In the past ten years, each bad agricultural cycle was followed by a positive one; as such producers have been able to compensate for the negative effects of the first cycle with a recovery the following year, without having to resort to agricultural insurance. The risk would increase if bad weather patterns were to last for two consecutive years, or longer, thus exhausting producers’ financial buffers. Another reason for producers’ reluctance to insure production is that they consider policy rates are too high to be profitable from their perspective. In the livestock sector an alternative to insurance are contractual guarantee clauses that include ranges of volumes and operational timeframes. For small producers, insurance is not a consideration, for them the biggest risk is credit risk in a year with a bad harvest, their priority at that point is the need for credit refinancing. Finally, many business agreements between companies who provide inputs and the production sectors are kept informally. Certainly no clauses or risk prevention elements are taken into consideration. Activity in the construction sector varies with the purchasing power stemming from agricultural GDP. Housing construction in particular adjusts with the agricultural cycle. This co-movement is also observed in other sectors that depend on Paraguayan purchasing power, for example, the hotel and restaurant sector, or wholesale and retail trade. Companies engaged in road construction have not been affected because they operate mainly as public sector contractors. 27 2.3 The impact of volatility originating in the agricultural sector on macroeconomic aggregates Shocks to agricultural GDP lead to a positive response in non-agricultural GDP and exports. The impact of agricultural volatility can be measured against the rest of the economy. Impulse responses on the basis of VAR analysis show that a shock to agricultural GDP translates into a response in non-agricultural GDP with a 16:1 ratio: this is for agriculture excluding cattle, fishery, and forestry, the effect is larger for the broader definition of agriculture (Berument 2013). A wavelet analysis of agricultural versus non-agricultural GDP shows that agricultural GDP leads non-agricultural GDP at wavelengths of 1 to 3 quarters. At higher wavelengths the direction of the lead is reversed, suggesting that in the medium-term overall development of the economy feeds back into fostering agricultural production (Figure 2.5). 27 Exports respond to agricultural GDP shocks with a ratio of 5:1 and again the effect is stronger for a broader definition of agriculture, in this case reaching 1:1 (see Berument (2013) for details). Figure 2. 5: Wavelet analysis of agriculture and non-agricultural GDP 3 0.8 4 0.6 6 Period 0.4 8 11 0.2 16 10 20 30 40 50 60 Time Source: Berument (2013a) The nominal exchange rate immediately absorbs fluctuations in agricultural export values, explaining the stability of the real exchange rate over the period analyzed. Export revenues of agricultural products are mainly in US dollars. As a consequence, a strong harvest generates dollar inflows and exercises upward pressure on the exchange rate. Paraguay has a managed float exchange rate regime whereby the Central Bank intervenes to avert abrupt changes. Exchange rate volatility has been increasing at the same time as the volatility of the agricultural sector (Annex 1.2). Market participants from sectors other than grain production note that the agricultural production cycle exerts an impact on the exchange rate, which renders operation in the export sector a challenge. 27 Berument (2013a) provides details on the analysis. 28 Figure 2. 6: Wavelet analysis of agriculture Figure 2.7: Wavelet analysis of non- and private consumption agriculture and private consumption 3 3 0.8 0.8 4 4 0.6 0.6 6 6 Period Period 0.4 0.4 8 8 11 11 0.2 0.2 16 16 10 20 30 40 50 60 10 20 30 40 50 60 Time Time Source: Berument (2013a) There is some indication that private consumption plays a role in propagating the impact of agricultural GDP through the economy, impacting non-agricultural GDP. Wavelet analysis shows that in the middle segment of the data set, agricultural activity seems to have induced consumption at wavelengths of 3 and 4. Toward the end of the sample, the evidence is either mixed or it does not lie within the cone of reliability (Figure 2.6). Even though this is not strong evidence it gives some indication of a relationship between agriculture and consumption. In turn, private consumption induces non-agricultural activity at wavelengths up to four quarters (Figure 2.7). Unemployment negatively correlates with agricultural GDP, in some sectors it is perceived to be directly impacted by agricultural GDP and fluctuates significantly with it. The unconditional correlation between unemployment and agricultural GDP amounts to -0.2. An in- depth quantitative analysis of the relationship is restricted by the lack of quarterly data on employment. Within agriculture, the cattle sector is relatively labor intensive and therefore employment in this subsector varies more with the production cycle than in the case of soy, which is more capital intensive. In terms of other sectors, the seed sector experiences strong fluctuations in hired personnel as a consequence of variation in agricultural production. Volatility in agriculture also impacts the public sector, through the effect that soy and beef exports exert on fiscal revenues. A positive and significant relationship can be established between soy and beef prices versus fiscal revenues, modeling the relationship in a two-step approach. Favaro, Koehler-Geib, Picarelli, and Indaco (2013) find that beef and soybean exports respond strongly to prices (using the canonical Nerlove (1959) model), they then find a positive and statistically significant relationship between tax revenue collection and the value of exported beef and soybean. A caveat to the analysis at the first step is that due to data restrictions export volumes instead of production volumes are used. The response in actual production may be lower than the estimated elasticities in this approach. The result of the second step is not trivial given the low direct taxation of the agricultural sector. The results seem to indicate that the positive relationship is due to value added tax. Beef and soybean production generate income 29 that is spent inside Paraguay for the most part. Part of this expenditure generates tax revenue via VAT and another part generates revenue through corporate income tax. The elasticity of soy exports to price changes exceeds that of beef, which could be linked to the limited time that soybean brokers have to hold the crop rather than commercialize it while there is more room for timing decisions in the case of beef. When it comes to the relationship between tax revenues versus soy and beef exports, the elasticity of revenues is higher in the case of beef. This is in line with how much more labor intensive beef is than soy and how it is more integrated into the value chain in Paraguay.28 Significant costs that overall GDP volatility imposes in terms of welfare, equity, and poverty have been established in the economic literature, and it is likely that this also applies to volatility originating in the agricultural sector. Poverty and distributional impacts of volatility were not the focus of the study. Lopez-Calva, Lugo, and Barriga Cabanilas (2013), forthcoming, are covering aspects of this topic. For developing countries, macroeconomic volatility, as summarized by output volatility, is reflected disproportionately in consumption volatility, and the welfare gains from reducing consumption volatility can be substantial (Loayza, Ranciere, Serven, and Ventura (2007)). Based on the approach of Athanasoulis and van Wincoop (2000), and World Bank (2000) estimated potential welfare gains of up to 5 to 10 percent of consumption in various Latin American countries. The negative link between macroeconomic volatility and equity has also been established in the literature.29 According to Breen and Garcia- Penalosa (2005) a country like Chile could reduce its Gini coefficient by 6 points if it were to reduce its volatility to the same level as Sweden or Norway. As argued in Lopez-Calva, Lugo, and Barriga Cabanilas (2013) forthcoming, a reason for the link between high volatility and inequity could be that citizens at the lower end of the income distribution have reduced access to insurance mechanisms and therefore suffer more from negative shocks. Macroeconomic volatility may also contribute to still elevated poverty rates; the high degree of volatility may be the weak link between solid average growth performance and employment generation. The uncertainty resulting from volatile economic growth may reduce the incentive for firms to employ new staff. Together, lagging employment generation and continued high levels of inequity pose important challenges for Paraguay in reducing poverty further. Simulations of a negative and persistent shock to beef and soy prices illustrate the existence of links between the agricultural sector and equity as well as poverty in Paraguay. A Computable General Equilibrium (CGE) model is used to track the macro and micro economic effects in Paraguay of a decrease of 25 percent in the soy and beef prices starting in 2013 and maintained through 2018 (see Diaz-Bonilla and Cicowiez (2013) for a detailed description of the model, the base line scenario and simulation results). In particular, a decrease in the world export prices of soy and beef would result in slower GDP growth than under the baseline scenario. Moreover, through a negative impact on the private sector (including reduced employment growth and private consumption), poverty would decrease to 25.8 percent in 2018 as opposed to 24.5 percent in the baseline simulation. Inequity and the aims of the millennium development 28 It seems to be important to take into account the indirect way in which commodity prices impact fiscal revenues, in a cointegration analysis of fiscal revenues versus beef and soy prices with yearly data from 1990 to 2010, Le Fort (2013) cannot detect a statistically significant relationship. 29 See for example Breen and Garcia-Penalosa (2004), Garcia-Penalosa and Turnovsky (2004) or Huang, Fang, and Miller (2012). 30 goals would remain practically unchanged. In the case of a general decrease in all of Paraguay’s exports , the impact would be much stronger, and would then include a negative impact on the millennium development goals and inequity 31 Chapter 3: Managing growth volatility in Paraguay Previous chapters have identified two main types of the sources of growth volatility in Paraguay: shocks of a macroeconomic nature and shocks specifically linked to agriculture. The main macroeconomic sources of volatility in Paraguay are shocks to global interest rates, foreign demand, terms of trade, investment, GDP itself, and pro-cyclical fiscal and monetary policies. Agricultural volatility is mainly driven by shocks to production such as rainfall; investment levels; and disease outbreaks and shocks to markets including commodity prices, the closing of markets in the case of disease outbreaks, and prices of imported inputs like fertilizers and pesticides. The purpose of the current chapter is to discuss policy options and tools to stabilize the economy by rendering it more resilient to the sources of volatility, and to mitigate the impact of volatility. Sources of volatility are interrelated and taking a broader perspective allows finding optimal ways to manage observed volatility and risks. Therefore, it is important to develop a comprehensive macroeconomic risk management framework that takes all different sources of volatility and risks into account and puts forward a coherent set of measures aimed at increasing Paraguay’s capacity to prepare for and cope with the effects of volatility. Any policy option needs to be assessed in terms of its fiscal implications; be it in terms of its effects on sustainability, on redistribution, or on potential contingent liabilities. In response to the concrete shocks that previous chapters have identified, the macroeconomic tool set in this chapter presents: i) an overall strategy with the aim of diversifying economy in a way that renders it less dependent on products and markets that introduce volatility; ii) policies that render factor markets more flexible; and iii) fiscal policies aimed at smoothing or at least avoiding amplification of shocks in the economy, through tools like fiscal rules or stabilization funds. The second set of tools contains measures of agricultural risk management designed to address the production and market shocks specific to the agricultural sector. Given that the suggested policy options are new and have only been applied in a few countries, this section relies on case studies. The idea is to provide some useful tools and experiences from other countries that have been facing similar volatility to that observed in Paraguay. However, a careful assessment of priorities among different options, in particular also their applicability, and their fit within the country’s comprehensive macroeconomic risk management framework is outstanding and needs to be part of an overall assessment of agricultural risks. The Government and the World Bank are currently collaborating on such an assessment, which aims at the adoption of an action plan of priority measures to mitigate, transfer and absorb risks affecting Paraguay’s agricultural sector.30 First, the agricultural risk management section presents four case studies on production risks: i) building animal health capacity to prevent foot and mouth disease in Colombia; ii) introducing weather derivatives based on a rainfall index for severe drought in Malawi; iii) establishing a weather contingency fund for the agricultural sector (CADENA) in Mexico; and iv) implementing an index-based livestock insurance project in Mongolia. Second, three further case 30 See World Bank (2013b) for a detailed description. 32 studies provide insights into new tools to address agricultural market risks: i) developing the asparagus market in Peru; ii) introducing subsidies for commodity price hedging contracts in Mexico; and iii) introducing agricultural commodity exchanges in Argentina. To complete the presentation of policy options for agricultural risk management traditional measures are presented in the appendices 3.1 and 3.2. It is important to develop a comprehensive macroeconomic risk management framework that takes all different sources of volatility and risks into account. Sources of volatility are interrelated and taking a broader perspective allows finding optimal ways to manage observed volatility and risks. Section 3.1 will describe the macro-economic toolbox to address growth volatility; section 3.2 will present the agricultural risk management toolbox mainly in the form of case studies; and section 3.3 concludes on a comprehensive macroeconomic risk management framework. 3.1 The macroeconomic toolbox to address growth volatility The purpose of the current section is to discuss policy options for addressing volatility, which arises from shocks to global interest rates, foreign demand, terms of trade, investment, GDP itself, and pro-cyclical fiscal and monetary policies. Section 1.2 identified these shocks as the main sources of volatility in Paraguay and chapter 2 discussed their impact on the agricultural sector and on the rest of the economy. The current section discusses policy options for Paraguay to address those shocks. It thereby complements the comprehensive analysis of “Managing Risks for Development,� World Development Report 2014 (World Bank (2014), forthcoming)). The recent World Development Report 2014 on Managing Risks for Development provides a useful analytical framework for developing an effective macroeconomic policy toolbox for Paraguay. The WDR discusses a number of country experiences with policies aimed at preparing for or coping with risks or shocks. Transferring risks through insurance mechanisms such as the use of sovereign bonds with pay-off structures associated with the occurrence of certain shocks is a third category of risk policies presented by the WDR. Having macroeconomic policies in place that safeguard macroeconomic stability is an important prerequisite for a country to face economic shocks. Consolidating Paraguay’s important progress on macroeconomic stabilization over the past years, in particular the control of inflation and the move to a flexible exchange rate regime is therefore a critical for its economy to face its increasingly volatile environment. Prudent fiscal policies in general, and specific fiscal policy instruments like fiscal rules are important policy ingredients for both preparing an economy to face shocks and coping with them – by providing the fiscal space to respond and avoiding amplification of the effects through pro-cyclical policies. A key priority for policies with a more medium and long term time horizon is to diversify the Paraguayan economy and strengthening those sectors that are less vulnerable to shocks. In the following, policy options for Paraguay along these lines are discussed in more detail, with a focus on policies aimed at diversifying the economy and fiscal responses. . The diversification of the economy and the development of domestic debt markets would reduce exposure to the global interest rate, foreign demand, and terms of trade shocks. As 33 discussed in chapter 1, the strong dependence on agriculture means that the economy is susceptible to interest rate shocks. The reform the agricultural corporate income tax, IMAGRO which has been realized in 2013 is an important first step. To level the playing field with other sectors, there is a need to eliminate all exemptions to ensure an appropriate taxation of the agricultural sector. Removing this distortion would level the playing field for business development in all sectors. A full evaluation of the recent reform is beyond the scope of the current study; however, it appears that further steps are needed to achieve an adequate taxation of agriculture. Also, addressing gaps in infrastructure, health, and education would contribute to an environment for entrepreneurs to explore new business opportunities. In addition, policies that contribute to a reduction in the concentration of the agricultural sector in terms of products and export destinations would reduce the exposure to foreign demand and terms of trade shocks. A diversification strategy for the agricultural sector would be appropriate. Promoting such diversification would require investment in human capital, access to credit and a fluid exchange of knowledge between entrepreneurs and universities and research centers. Export promotion activities would contribute to exploring new export markets for Paraguayan products. There are several promising policy options forpromoting a broader growth pattern by enhancing the regulatory and policy framework for all sectors. This includes measures that: i) improve factor market flexibility; ii) facilitate innovation and its application, and iii) to improve forecasting of future economic activity. As discussed in section 1.3, factor markets still show significant inflexibilities and rigidities. There are several entry points to addressing these rigidities:  Strengthening the domestic financial market, such as by improving access to credit for small firms.  Updating the legal framework for business activities with a view to reducing barriers to inter-sectoral factor mobility.  Improving the flexibility of the labor market, by rendering regulations and improving the education system with a view towards labor market needs (workers with a solid educational background are more fungible and switch jobs more easily).  Improving economic forecasting in Paraguay by linking universities and research centers to the business sector and thereby fostering a fluid exchange of knowledge. This would increase the predictability of the business environment and improve the knowledge base in the business community.  Finally, market participants mention the lack of information about the patterns and the effects of volatility as a major challenge to operating in the Paraguayan economy. In this context an important step would be to expand and improve statistics and data on weather conditions as well as a coordinated analysis. Dependence on international interest rates could be reduced through the development of the domestic debt market and increased use of public pension funds for domestic investments. (see World Bank (2013) and World Bank and IMF (2012) for a detailed description of the necessary steps). An additional measure for reducing the exposure to international interest rate shocks consists of ensuring that the funds of the public pension system (amounting to approximately 2 percent GDP) can be invested in the country and increase liquidity instead of being locked in an account at the Central Bank. Policy instruments like fiscal rules and stabilization funds that could address the observed pro- cyclicality of fiscal policy in Paraguay and help create fiscal space to mitigate the effects of volatility. As stated in section 3.1, fiscal policy in Paraguay has been pro-cyclical in the last two 34 decades with a few exceptions in recent years. Experiences from other countries have shown that policy instruments like fiscal rules and stabilization funds can help avoid such pro-cyclical effects. These instruments are no panacea; however, as their effective implementation depends on their credibility, which in turn is a function of the ability and incentives of the political decision makers to circumvent them. The design of fiscal rules and stabilization funds, but also the general environment, such as the existence of a broad political consensus, an adequate level of accountability and transparency in political processes determine their level of credibility. Fiscal rules are institutional mechanisms aimed at supporting fiscal discipline and attaining sustainability of public debt, control of public spending, and contribution to cyclical stability . There are two major categories of fiscal rules: one category defines numerical targets (i.e., ceilings or floors) for Government balances, overall revenues or expenditures that are fixed and independent of the business cycle (i.e. the Stability and Growth Pact in the Euro Area), and the second category aims at stabilizing cyclically-adjusted balances, allowing for cyclical changes in actual Government balances. Numerical targets are easier to communicate, and to verify by market participants. Structural budget balances have the advantage of providing short term flexibility to respond to adverse shocks. They are vulnerable though to uncertainty over the cyclical position of the economy and to over-optimistic GDP growth and budget forecasts. Stabilization funds are designed to guard against volatility in the international markets and aim to reduce the impact of volatile revenue on the fiscal balance and the economy. The basic concept behind stabilization funds is that when revenues and prices are high, windfall gains are diverted as payments into the stabilization fund. When revenues are lower than expected, payments are made out of the fund to the budget to avoid a sudden fall in expenditure. Stabilization funds are usually implemented in resource-rich countries that rely heavily on one or a few commodities for their fiscal revenues. Chile is a successful example of a country that has adopted a fiscal rule based on a cyclically adjusted fiscal balance. In 2001, Chile adopted a cyclically adjusted Government balance rule which links Government spending to cyclically-adjusted revenue, taking into account cycles in GDP and mineral prices. Among the 10 countries using fiscal rules based on cyclically adjusted fiscal balances, Chile is the only country that corrects not only for the cyclical deviation of GDP from its trend, but also for those of copper prices from trend. An important and innovati ve feature of Chile’s fiscal framework is the determination of GDP and copper price forecasting to two independent committee, whose projections are a legally binding input into the application of the fiscal rule. Adopting the fiscal rule has contributed to lowering the pro-cyclical bias of fiscal policy in Chile and has stabilized its macroeconomic environment. Namely, fiscal sustainability and credibility have been increased, the sovereign risk premium and macroeconomic uncertainty have dropped, and the volatility of GDP, interest rates, and the exchange rate have been reduced. Moreover, the dependence on foreign financing during downturns has been be reduced (Schmidt-Hebbel, 2012). Adopting and effectively implementing a fiscal rule in Paraguay would require a series of fiscal policy reforms. Difficulties in the practical application of fiscal rules have to be taken into account and translated into a pragmatic approach that is tailored for the specific situation in Paraguay. IMF (2009a) and Debrun, Hauner, and Kumar (2009) provide more details on the preconditions for the successful implementation of fiscal rules. The recent introduction of a fiscal responsibility law with the aim of strengthening the fiscal policy framework is a critical first step in this regard. An in depth analysis of the impact of the recent reform goes beyond the scope of this study but would be warranted given the potential impact on the economy. Overall, fiscal rules with fiscal responsibility 35 laws are more difficult to reverse, although it can take longer to establish them when economic and political uncertainty exists in a country. This will have to be complemented by other reform steps:  Adequate public financial management systems are prerequisites for effective implementation of fiscal rules. In Paraguay a careful evaluation of these systems would have to precede further steps.  An independent fiscal council could help in the formulation and implementation of sound fiscal policies. In particular, a fiscal council can complement the role played by existing institutions and enhance the effectiveness of fiscal rules.  Fiscal rules also need to include accountability, transparency, monitoring, external control, auditing, and enforcement mechanisms. Along with Chile, and Mexico, Norway is an example of a country that has managed revenue volatility through the implementation of a stabilization fund. Established in 1990 and activated in 1995, Norway’s Stabilization State Petroleum Fund (SPF) is designed to manage accumulated budgetary surpluses from oil revenues and has flexible operation principles with no specific rules for accumulation or withdrawal. The SPF effectively finances the overall budget balance by transferring net oil revenues from the budget to the SPF and in turn, financing the budget’s non-oil deficit through a reverse transfer. In addition, an overall budget surplus will be transferred to the fund and a budget deficit is financed by the fund. The accumulation of assets in the SPF, which include the returns on the fund's capital, represents Government net financial saving. The amount actually saved depends on oil prices and the fiscal outturn that contains the non-oil fiscal deficit. Controlled by the ministry of finance and managed by the central bank, the SPF assets have a high level of transparency and accountability. The size of accumulated funds reached close to 20 percent of GDP at end-1999 and has been increasing rapidly. Chile’s Copper Stabilization Fund (CSF) has helped the Government resist expenditure pressures during the increases in copper prices, reducing the cyclicality of fiscal policy. Established in 1985 following a sustained increase in the international copper price, the CSF's accumulation and withdrawal rules are based on a reference copper price determined annually by the authorities.31 The resources of the CSF have grown substantially since 1987, although in 1998-99 there were significant withdrawals, partly on account of a sharp decline in copper prices. In recent years, CSF resources have been used to subsidize domestic gasoline prices through credits to the Oil Stabilization Fund. The establishment of the CSF has allowed the Government to resist expenditure pressures during increases in the copper prices in the late 1980s and mid-1990s and to escape pro-cyclical fiscal policy. Davis, Ossowski, Daniel, and Barnett (2001) found a negative correlation between a copper price increase and Government spending. Sound fiscal and macroeconomic policy in Chile seems to have played a key role in helping the effective implementation of the CSF. Through a new Fiscal Responsibility Law enacted in 2006, Chile has modernized its stabilization fund and strengthened its link to the overall fiscal policy framework. The Law created two sovereign wealth funds: (a) the new Pension Reserve Fund (PRF), created to finance future pension 31 No explicit formula is used to calculate the reference price. In practice, however, the reference price followed a ten-year moving average until the mid-1990s; more recently, the reference price has been set somewhat lower than the moving average. When the price of copper exceeds the reference price by between $0.04 and $0.06 a pound, 50 percent of the resulting state copper company's revenues is deposited in the CSF; above $0.06 per pound, 100 percent. The rules for withdrawals are symmetric (OECD, 2009). 36 liabilities by the government. (b) the Economic and Social Stabilization Fund (ESSF). The Law established that, in good times, fiscal surpluses in excess of the structural target (and after contribution to the PRF) are channeled to the ESSF. In bad times, resources may be withdrawn from ESSF to finance budget deficits, including payments into the PRF.32 Mexico’s Oil Stabilization Funds are an example that illustrates that stabilization funds can produce limited results due to excess revenues allocation rules and capped savings. In order to reduce oil-related volatility in the budget, Mexico established three oil revenue stabilization funds: one by the Federal Government, a second one by the state-owned petrol company (PEMEX), and a third one by the State Governments. The first one was established in 2000 and the other two in 2006. The rules of the funds were updated in the 2006 Fiscal Responsibility Law and in the 2009 budget.33 The Federal Government fund is managed by the Ministry of Finance and has a target level for savings, which was 0.5 percent of GDP in 2008 and was almost doubled in the budget for 2009.34 As determined by law, 90 percent of excess revenues are allocated to those three funds (40 percent to the Federal Government fund, and 25 percent to the PEMEX and State Government funds each) and the remaining 10 percent to states for investment. Once the funds have reached their limit, 75 percent of excess revenues are allocated to investment, and 25 percent to a fund to support the restructuring of pension systems. At end-2008, the funds’ cumulative reserves were equivalent to 1.2 percent of GDP. Due to the cap on their size, the Mexican funds have accumulated a limited amount of savings and have therefore showed limited success in reducing volatility (OECD, 2009). With the adoption of the fiscal responsibility law, a pre-condition for establishing a stabilization fund in Paraguay has been fulfilled, but other conditions, such as the creation of an advisory committee would need to be put in place. A functioning stabilization fund requires a number of conditions to be in place: i) effective and transparent corporate governance; ii) transparent information of the transfers between the budget and the stabilization fund; iii) portfolio composition determined by maturity concerns (determined by the length of commodity- price and output cycles) and the Government’s degree of risk aversion, and iv) efficient portfolio management using transparent guidelines and closely monitored by the Government and the public, independent of political consideration (Schmidt-Hebbel, 2012). Other critical conditions for the effective operation and implementation of a stabilization fund include the adaptation of legislation and institutions that define investment policies and management principles of their funds. The status of these preconditions would have to be carefully assessed in the case of Paraguay. Establishing a stabilization fund is most effective in combination with the introduction of a fiscal rule. One important factor that strengthens the effectiveness of stabilization funds is the introduction of fiscal rules. Countries can establish stabilization funds with and without fiscal rules for expenditure and revenue smoothing. However, adopting fiscal rules when stabilization funds are established is critical; without fiscal rules regarding liquidity constraints, stabilization 32 Schmidt-Hebbel (2002) documents in detail the institutional aspects of fiscal policy in Chile and compares them with those of Norway. 33 The Law also included provisions for setting a reference price for oil and transfers to the funds. 34 Before transferring excess revenues to the funds, some items are deducted, which include shortfalls in revenues with respect to the budget, changes in energy costs that are not fully reflected in domestic electricity tariffs, costs of natural disasters and outlays resulting from changes in non-programmable expenditures due to changes in interest or exchange rates (OECD, 2009). 37 funds are unable to stabilize expenditure directly and Governments could finance spending through borrowing bypassing the operations of the stabilization fund. Expenditure smoothing therefore requires additional fiscal policy decisions besides the operation of the fund. 3.2 The agricultural risk management toolbox The purpose of this section is to present some useful tools and experiences from other countries that have been managing volatility similar to that observed in Paraguay. First, the section introduces the agricultural risk management framework used by the World Bank. It then presents four case studies on new tools and approaches to mitigate, cope with, and transfer agricultural production risks: i) building animal health capacity to prevent foot and mouth disease in Colombia; ii) introducing weather derivatives based on a rainfall index for severe drought in Malawi; iii) establishing a weather contingency fund for the agricultural sector (CADENA) in Mexico; and iv) implementing an index-based livestock insurance project in Mongolia. Second, three case studies provide examples of measures to mitigate and transfer agricultural market risks: i) developing the asparagus market in Peru; ii) introducing subsidies for commodity price hedging contracts in Mexico; and iii) introducing agricultural commodity exchanges in Argentina. To complete the presentation of policy options for agricultural risk management traditional measures of risk management are presented in the appendix to the chapter. Each case study highlights its relevance to Paraguay and indicates benefits and limitations associated with the given approach. The case studies identify directions to guide further research to determine whether the program is appropriate for Paraguay. However, a careful assessment of priorities among different options, their applicability, and their fit within the country’s comprehensive macroeconomic risk management framework is outstanding and needs to be part of an overall assessment of agricultural risks. Case studies cannot be applied immediately to Paraguay. A careful assessment of viable policy options is provided in World Bank (2013b). For all new policies and programs in agricultural risk management, an informed decision-making process relies on a sector-wide risk assessment to identify hazards, vulnerability, and exposure to risk, followed by cost-benefit analyses to weigh different options. It is also important to link it to the overall macroeconomic environment because the suggested solutions may have implication on fiscal sustainability, redistribution, and contingent liabilities for the Government. Overall, the objective of an explicit agriculture risk management strategy as part of a comprehensive macroeconomic risk management framework is to move from ad-hoc, ex- post responses to adverse shocks to agriculture, to the establishment of an ex-ante risk management framework. This allows the Government to better manage fiscal exposure (revenues and/or expenditures) in case of systemic shocks to agricultural production. In recent years, the Bank has developed a framework for supporting Governments in defining their agriculture risk management strategy. This framework is described below. This strategy fits well into a broader set of fiscal policy tools to manage volatility that may arise from fluctuations in agricultural GDP, such as the establishment of a fiscal rule and stabilization funds. 38 Agricultural GDP volatility can derive from risks associated with production, market, and the enabling environment. For the agriculture sector of Paraguay, production risk and market risk are the most important sources of risk: i) production risks arise from rainfall, investment levels; and disease outbreaks; ii) market risks arise from fluctuations in the prices of export commodities like soy, beef, and maize; fluctuations in the prices of imported inputs; the closing of markets (such as the border closings due to foot and mouth disease outbreaks); and volatility in the prices of imported inputs like fertilizers and pesticides. There are three main strategies that comprise an integrated agricultural risk management strategy and the case studies of this section are categorized accordingly (Figure 3.1): i) mitigation: activities designed to reduce the likelihood of an adverse event or reduce the severity of actual losses (e.g. diversification, animal and plant health investments; ii) transfer: the transfer of the potential financial consequences of particular risks from one party to another, for a fee or premium (e.g. commercial insurance and hedging); iii) coping: improves resilience to cope with (respond to) events, through ex-ante preparation (e.g., social safety net programs, buffer funds, savings, strategic reserves, contingent financing, etc.) Figure 3.1: The World Bank Agricultural Risk Management Framework Source: World Bank (2013a). Strategy Problem Instrument Description Case Study Foot and mouth Public-private Joint PPP for investing in Colombia: (foot and mouth partnership (PPP) sanitary measures, standards Building animal disease) outbreak in investments in regulation, certification, etc. to health capacity to the Andean Region Animal Health mitigate risk of disease prevent foot and Mitigation threatens the outbreak and remain free of mouth disease and important cattle/beef foot and mouth disease in order support the export sector of to maintain access to export livestock sector Colombia. markets. 39 Severe droughts Public sector Financial contract (derivative) Malawi: Weather pose significant purchase of an by which payment to the derivative based on food insecurity Index-based weather Malawi Government is rainfall index for problems for the derivative to gain provided when rainfall in a pre- severe drought Transfer vulnerable access to quick and specified period falls within a population segments appropriate level of pre-specified threshold. With of Malawi. resources to the payout, the Government respond. purchases food aid. Severe weather Federal and Emergency fund financed by Mexico: Weather events make it subnational Government savings and index- Contingency Fund difficult for small Governments based insurance that provides for the Agriculture farmers to invest purchase Index- direct payments to small Sector (CADENA) and exist the based insurance to farmers in a given municipality Transfer poverty cycle. obtain additional affected by catastrophic fiscal resources to weather event. compensate farmers after an adverse weather event Harsh winters force Contingency lines of Provides fiscal resources after a Mongolia: Index- the Government to credit to fund harsh winter (dzud) in order for based Livestock respond with aid to emergency response the Government to make Insurance Project Coping low-income herders activities and catastrophic payments and that have lost a large payments. provide assistance to herders. amount of livestock Source: authors. Instrument: Public-private investments in animal and plant health ARM Strategy: Mitigation Relevance for Paraguay: In 2011 an outbreak of foot and mouth disease in Paraguay led to the mandatory slaughter of 1,000 head of cattle. Further outbreaks have been reported since then (January 2012). Paraguay’s status as free of foot and mouth disease with vaccination has been suspended by the OIE. 35 Chile, which had previously purchased roughly a third of Paraguayan beef exports, banned Paraguayan beef. Total beef exports dropped 16.5 percent in 2011. Paraguay is strengthening investments in a sanitary and phytosanitary (SPS) system and is establishing a national biosafety laboratory level 3. Colombia’s efforts to strengthen the National Agricultural Science and Technology and Sanitary and Phytosanitary (SPS) systems via public and private sector participation improved the access of Colombia’s export products to international markets. Colombian agricultural and agro-entrepreneurial sectors accounted for 21 percent of aggregate GDP, 25 percent of export revenues, and 30 percent of job creation in the country, employing more than 4.5 million. The livestock sector was extremely vulnerable to lapses in quality standards. Venezuela, the principal market for Colombian beef, closed its border with Colombia and sent an 35 World Organization for Animal Health (2013). 40 extreme shock through the sector. Furthermore, on the brink of joining a Free Trade Agreement with the United States, strengthening SPS standards was a necessary step to ensure competitiveness in international markets. The SPS strengthening strategy involved national disease-free certification, low-tech implementation of good agricultural practices (GAP), and the approval of export protocols with many countries. The country was certified free of foot and mouth disease without vaccination, and several plant and animal disease-free areas were established (among the most important, Brucelosis, Tuberculosis, Bactrocera, controlled fruit fly). With respect to the eradication of foot and mouth disease, the country has complied with the commitments of the Hemispheric Plan of Eradication of Foot and Mouth Disease (PHEFA). To maintain foot and mouth disease-free status by the OIE and PHEFA, Colombia decided to establish an in-country Biosafety Level 3 Agriculture Laboratory. As part of the strengthening of the SPS laboratory network, Colombia needed to respond to the rising threat of foot and mouth disease to the livestock industry. Such a laboratory serves an important role in a prevention system by analyzing samples and monitoring standards control. The investment was justified from the point of view of the large returns to the local livestock industry (local consumptions and exports), but also for the Region. This laboratory is the only one of its level of biosafety in the Andean Region. The availability of a national biosafety level 3 agriculture laboratory for Colombia and for the Andean Region is a resource that can yield large economic returns by allowing for early and precise surveillance, control and monitoring of exotic or emerging animal health issues. Table 3. 2 Colombia’s study case. Benefits, Challenges and Considerations for Paraguay Benefits Challenges Considerations for Paraguay  Disease free  Coordination between  Disease prevention reduces livestock loss, certification is private and public actors requisite slaughter of sick animals, and increases mandatory for export and clear definition of roles access to export markets markets with higher  Integration between  Multiple outbreaks of foot and mouth disease quality standards components indicate need for further  Early and precise  Up-front costs of intervention/investments. Interventions must be surveillance of animal establishing a new specified to target the unique gaps and and plant disease laboratory; operational weaknesses in the existing system. An analysis  Multi-faceted approach costs of collecting and to determine the causes of outbreaks is combines low- analyzing samples necessary to guide future investments. technology extension  Targeting investments for  Paraguay is in the process of establishing a for good agricultural cost-effectiveness similarly advanced certified laboratory. practices with advanced  To regain foot and mouth disease-free national technology status, coordination with the OIE is necessary. What would the gains to Paraguay be if it were to regain foot and mouth disease-free national status? How sensitive are current consumers of Paraguayan beef to foot and mouth disease concerns? Source: authors. 41 Instrument: Federal and state Governments purchase index-based insurance to obtain additional fiscal resources to compensate farmers after an adverse weather event. ARM Strategy: Transfer Relevance for Paraguay: A majority – 83.5 percent (nearly 242,000) – of Paraguay’s farms are less than 20 hectares. Such smallholder farmers cannot qualify for commercial agricultural insurance. Index insurance has a number of advantages over traditional insurance and traditional disaster response programs for covering small farmers, but there are numerous difficulties in implementation at the same time. CADENA (Componente Atencion a Desastres Naturales en el Sector Agropecuario y Pesquero) is a macro-level catastrophe crop and livestock insurance program that is specifically designed to provide a social safety net for vulnerable smallholder farmers that do not qualify for commercial agriculture insurance. The CADENA program is designed to replace the Government’s traditional ad-hoc disaster relief schemes. Instead, States purchase parametric crop and livestock insurance to cover a pre-registered rural population, which receive automatic payments in the case of a catastrophic disaster, regardless of their individual, farm- level losses. CADENA is designed to quickly provide income-compensation to smallholder farmers to help them recover from a catastrophic event and continue production. Under CADENA index insurance, farmers are not compensated for their actual losses and instead receive payments based on whether their location was affected by a disaster. CADENA actively promotes pre-registration of farmers so that payments are fast and transparent. Under CADENA, smallholders do not pay any part of the premium. Rather, the Ministry of Agriculture subsidizes either 80 or 90 percent of the insurance premiums, depending on the degree of marginalization of farmers in the state, and the State Government pays the remainder. Beneficiaries are eligible if they meet certain criteria for smallholder producers in terms of size of property, number of livestock, etc. Mexican states are incentivized through federal Government premium subsidies to contract agricultural insurance. States can either directly contract insurance from a private insurer or Agroasemex, sharing the cost of the premiums with the Ministry of Agriculture in the proportions indicated above; or if they decline to contract insurance cover, States can still benefit from CADENA’s Direct Support program in the event of a catastrophe, but the state must shoulder 50 percent of the cost of the total estimated damages, with the Ministry of Agriculture compensating 50 percent of the costs. If a State declines insurance coverage, the Ministry of Agriculture is entitled to purchase insurance cover and pay 100 percent of the premiums, exclusively using insurance from Agroasemex, the public re-insurance company in order to hedge their exposure in case they need to provide direct payments. Mexican states must choose between parametric/weather index insurance from Agroasemex and area-based yield index insurance (AYII). Agroasemex is the only company that offers parametric weather insurance products. Private insurers only offer AYII. Neither form of insurance is indemnity-based, meaning that producers are not individually compensated for the specific quantity of damages incurred on their farm. The difference between 42 parametric/weather index insurance and AYII is that payouts from a parametric insurance are triggered by a pre-established weather variable that is correlated with agricultural losses, while AYII requires actual in-field sampling of crop yields to establish the actual average municipality- level yield loss. The parametric weather index covers a restricted number of risks while AYII covers multiple risks, including natural, climatic, and biological causes of crop production or yield loss. The two kinds of livestock insurance available are a parametric remote sensing pasture index, using a NDVI and traditional catastrophe livestock insurance. Since its inception in 2003 the CADENA program has expanded fast in terms of coverage and budget allocation. In 2011 approximately 8 million hectares of crops were insured in 27 states with over 2.5 million insured farmers (beneficiaries). This represents about 56 percent of this target group (4.5 million subsistence smallholders farming 16.5 million hectares). Overall the CADENA crop and livestock insurance programs in 2011 covered 2,362 municipalities in 30 out of Mexico’s 32 states36 with Total Sum Insured (TSI) of 12 billion (Ministry of Agriculture 2012). The Federal Government’s CADENA budget has increased significantly through the Ministry of Agriculture for support to catastrophe crop insurance premiums and direct compensation payments. In 2012 it reached US$ 232.7 million, of which 153.6 million (66 percent of total) was allocated to premium subsidies and the remainder of US$ 80 million for direct payments. For 2013, the Ministry of Agriculture has therefore significantly increased the federal Government financial budget for the CADENA Program to about US$ 400 million (representing an increase of about 72 percent on the 2012 budget). Weather index insurance structured like CADENA avoids many problems of traditional insurance, including: i) adverse selection: All farmers in a given region that qualify are automatically opted in to the insurance product. In the CADENA case, farmers do not pay directly for the premiums to the insurance so there is no willingness-to-pay obstacle; ii) moral hazard: farmers still have an incentive to try to save their crops, as the indemnity payout is perceived as an additional bonus regardless of actual losses; iii) high correlated risks: natural disasters typically strike entire communities, wiping out local coping mechanisms such as informal lending within a community; iv) transaction costs: index insurance can reduce or eliminate the need for in-field damage assessments.37 Traditional multiple or single peril crop insurance relies on surveys of field damage to determine the appropriate indemnity payment. CADENA also has several advantages compared to an ex-post disaster compensation program, like its precursor program: i) insurance payouts can be made rapidly to State Governments, and State Governments have some degree of autonomy over how to allocate resources in the case of a disaster; ii) insurance payouts can be made rapidly to farmers where there is an ex-ante farmer registry; iii) Transparency and standardization of payout rules; iv) subsidies for a public-private partnership may be less of a fiscal burden, and at the very least a more consistent fiscal burden, than an ex-post program; v) index insurance makes it possible to layer risk and enable risk transfer (reinsurance in this case). The maximum liability can be quantified in advance and transferred out of the fiscal budget to local and international insurance and reinsurance markets. 36 Mexico has 31 states plus 1 District Federal and a total of 2,445 municipalities. 37 See annex 3.2 for a comparison of principle agricultural insurance products. 43 Challenges for CADENA consist of the ensuring that the state distributes payments quickly to affected farmers in insured locations and to address the high basis risk. An external evaluation by the Universidad de Chapingo found that the average time post-event is 89 days for beneficiaries to receive payouts.38 Furthermore, CADENA has had difficulties monitoring how the state has transferred payouts or used the resources. High basis risk, the difference between the value of the insurance payout and the value of the beneficiary farmer’s actual loss, is large for index insurance. For many farmers, CADENA payouts are inadequate to cover their costs invested in agricultural production. Table 3.3: Mexico’s study case. Benefits, challenges and considerations for Paraguay Benefits Challenges Considerations for Paraguay  Avoids many limitations of  High basis risk  Adequate insurance market: Is such a traditional/commercial  Difficulties in implementation program feasible given the current insurance products (adverse (speed and transparency of technical level of local insurance selection, correlated risk, indemnity payouts) companies? Is there technical moral hazard, transaction  Indemnity payouts do not expertise in the market to offer index- costs) completely cover production based, low-cost insurance?  Advantageous over an ex- costs and instead serve to help  Issues with data and implementation: post emergency fund get farmers “back in Is there sufficient weather data (private sector contributes to production.� information to design a macro-level cost-sharing, designed to  PPP challenges: imperfect agriculture insurance product? increase speed of payouts, competition between public Registration of farmers may be possible to transfer risk insurance agency and private difficult or given widespread land instead of retaining all risk sector tenure insecurity. in the fiscal budget)  Paraguay’s natural hazards: Does the frequency and severity of natural hazards in Paraguay justify the transfer of such risks through insurance, or absorbing and diversifying the risk is more viable?  Integration with existing policies: How would such a program interact with other social safety net programs for small farmers and rural households in place? How can it be linked to or replace the current system for coping with disasters? What would be the fiscal burden comparison between state subsidies for premiums and a state emergency fund?  Would land tenure insecurity complicate farmer registration and indemnity payouts? Source: authors. 38 Universidad de Chapingo, External Evaluation to the PACC, 2010. 44 Instrument: Fiscal risk management via index-based weather derivative delivers timely and guaranteed contingent funds in case of emergency ARM Strategy: Transfer Relevance for Paraguay: Paraguay faces low frequency, severe production risks like drought that necessitate occasional large and urgent Government expenditures. In January 2012, the Government declared a state of food emergency in southeastern Paraguay due to drought, and the Ministry of Agriculture estimated that 30 to 50 percent of agricultural production would be lost.39 Food insecurity was exacerbated by difficulties in transport of food, as low water levels limited commercial shipping along Paraguay’s rivers and canals. Even though Paraguay is a net food-exporting country, like Malawi, it has a high number of farmers that rely on their own production for food security and faces food shortages and pressure for emergency responses. Malawi’s index-based weather derivative transfers the financial risk of severe and catastrophic national drought to the international risk markets with the World Bank as intermediary. Malawi has a high exposure to the risk of drought and food shortage. For a food- importing country with a high portion of the population dependent on agriculture, Malawi faced widespread hunger in 2005 when a severe drought struck. Millions of farmers needed food aid. The Government of Malawi spent $200 million responding to the crisis and donors contributed similar funds. Instead of waiting for international relief funds to mobilize, the Government of Malawi receives a payout from the World Bank if the index hits the pre-determined trigger. The derivative gave nearly immediate access to Malawi to funds to respond to the crisis, thereby reducing the country’s dependence on humanitarian aid. Weather-risk management transactions can be customized according to countries’ specific needs, the type of weather hazard, level of protection, and estimated financial loss associated with a severe and catastrophic event. Drought can be predicted and yield loss correlates closely with rainfall in the case of Malawi. The Government of Malawi has stopped purchasing a derivative but is now considering financing through international financial institutions including a draw down option. Pre-requisites for a weather derivative contract: i) index: an index that dependably captures national hazard (e.g. drought) risk; ii) data: high quality historical weather data and reliable real- time communication; iii) premium: an annual, non-refundable premium must be paid by the “insured� party or a donor; iv) integration: into a larger risk-management strategy. Table 3.4: Malawi’s study case. Benefits, challenges and considerations for Paraguay Benefits Challenges Considerations for Paraguay  Payout is timely and  Basis Risk: the potential  Which catastrophic events is guaranteed in time of need mismatch between the Paraguay most vulnerable to, and because it is index-based and contract payout and the what are the current coping is independent of actual actual maize production measures? production assessments losses whereas the payout  Does Paraguay face comparably 39 USDA (2012). 45  Creates opportunities to access does not adequately severe and frequent risks? Will cost- the market for risk transfer. indemnify the Government savings from weather derivative Systemic risk can be for losses. The index also justify this choice of instrument? transferred from a low-income only covers losses from a Malawi counted on donors to help country to investors. certain pre-specified shock. finance the premium for the weather  Cost savings: through early, Indexed risks: the contract derivative; can Paraguay garner such more efficient and planned only covers risks that can support? response to weather shocks be indexed – not other  Does Paraguay have availability of due to predictable crisis natural and man-made historic weather data to build index? financing. risks to food production.  Paraguay is a larger country than  Strengthens Government’s Setting up an index Malawi and may be better suited to ability to finance responses to requires historical crop and diversify risks across Departments or natural disasters, reducing the weather data and an sectors, rather than transfer them. country’s reliance on adequate network of  What are other, less costly measures humanitarian emergency weather data stations. to ensure food security? appeals.  Premium: these  Less fiscal volatility via transactions have an improved budget planning upfront cost. Source: authors. Instrument: Contingency lines of credit to fund emergency response activities and payments, based on livestock mortality index insurance. ARM Strategy: Transfer Relevance to Paraguay: Paraguay livestock production is concentrated in the Chaco, a region that faces high exposure to weather shocks like drought. In Mongolia, harsh winters occur roughly once every five years, killing millions of livestock and devastating the basis of the livelihood for nearly half Mongolia’s population. Roughly a third of aggregate GDP derives from the agriculture sector, of which nearly 80 percent comes from herding. The rural population relies heavily on livestock for income, employment, food security, and a means to invest wealth. Recent dzud events occurred in December 2009 and January and February 2010. Beginning in 2006, the World Bank helped the Government of Mongolia develop the Index- Based Livestock Insurance Program (IBLIP), which is a combination of self-insurance, market-based insurance, and social safety net. Layers of risk are allocated to different actors depending on severity. Herders assume small, frequent losses. Larger losses are transferred to the private insurance industry, for which herders pay a market premium rate. The Government of Mongolia bears the cost for the catastrophic loss risk layer. Since the project’s inception, insurance policies have become more and more popular among herders. Increasing numbers of farmers are purchasing insurance. After the first phase of the project in 2010, over 14,000 insurance policies had been sold. In 2009, indemnity payments were made to all 2,117 herders who were eligible following livestock losses. Local insurance firms remain committed to selling the product. 46 As an index-insurance product, insurance payouts do not compensate for individual livestock losses, but rather are triggered for a micro-region when the livestock mortality rate in the region exceeds a specific threshold. Good data makes this product possible. Since the insurance is not linked to the dzud event, the program relies on Mongolia’s three decades of time-series data on animal mortality per micro-region and for all species of livestock. After a specified “exhaustion point� that varies based on species and location, insurance companies are not liable and the Government financed and operated safety net program is mobilized. Index-based mortality insurance was chosen for its relative simplicity, low cost, and low risk of moral hazard and adverse selection. Alternatives considered include individual insurance coverage to herders and index-based weather insurance. Individual coverage has not been successful in Mongolia due to moral hazard, adverse selection, high administration costs, and an immature private insurance market. Index-based weather insurance was also considered but Mongolia does not have the historical weather data necessary to design a weather index. The dzud events themselves are also complex phenomena that are influenced by summer rainfall, winter snowfall, temperature, and wind. The Government of Mongolia was able to turn to international markets with a Contingent Debt Facility to finance these risks. By pooling risk, the Government of Mongolia could obtain global reinsurance on the pool. Such a contingency line of credit funds the Government’s emergency response. This is considered to be a more efficient way to provide subsidy. Furthermore, the partnership with the private insurance sector makes it possible for the insurance to stand on its own. If the Government of Mongolia decides to end the subsidy, the livestock risk insurance can still be sold. Table 3.5: Mongolia’s study case. Benefits, challenges and considerations for Paraguay Benefits Challenges Considerations for Paraguay  Risk layering  Willingness to pay  What incentives to purchase insurance  Relatively simple  Willingness to pay: given the risk do farmers (in particular smaller ones)  Reduced risk of layering approach, farmers pay for face? Does insurance increase access to moral hazard and small, frequent risks but are credit in Paraguay? What is the adverse selection covered for larger losses. willingness to pay of Paraguayan  Ex-ante budget Substantial outreach was necessary farmers for such a product? planning to reduce to educate about this new program  Same insurable asset across territory: fiscal exposure to and encourage farmers to purchase Mongolia’s reliance on livestock emergency events the insurance. production is unique. To have a large  Promote good  Domestic insurance market: The number of producers in different regions management domestic insurance market is very across the country with the same practices for herders small and highly concentrated, insurable asset (livestock) facilitated risk  Program was piloted with the largest insurance company pooling. In contrast, the majority of to test several at a market share of 74 percent. livestock raised in Paraguay is on large hypotheses before The IBLIP invested in capacity ranches and does not have the same full scale building and eventually helped significance to rural livelihoods. implementation develop the market and encourage  State of the insurance and reinsurance new insurance companies to enter market; would similar risk layering be and stronger products. appropriate, or do farmers prefer other  Significant outreach necessary to types of coverage? educate about insurance project  Is saving/borrowing the appropriate tool  Significant capacity building for the public sector fiscal management necessary in the insurance sector (rather than transfer of risks) given the 47  Premium on transferring the risk at scale of catastrophic events in Paraguay? the catastrophic layer Source: authors. Table 3.6: Instruments for Managing Market Risk Strategy Problem Instrument Description Case Study Mitigate Peru’s exports have Diversification of Historically dependent on Peru: been historically Agriculture Sector traditional exports of raw Development concentrated in few materials, Peru diversified into of the commodities (minerals high-value non-traditional Asparagus and fishmeal), agricultural exports through public Market influencing private partnerships. Now the agricultural GDP world’s leading producer of volatility asparagus. Transfer Problems with Subsidies for Subsidize premiums on options Mexico: enforcing forward commodity price contracts bought in international AxC contracts and domestic risk hedging markets in order to encourage price formation. contracts physical forward contracts. Transfer Reduce income Development of Develop local commodity Argentina volatility for growing Agriculture exchange to offer local (ROFEX and agriculture supply Commodity futures/options contracts accessible MATBA chains. Exchange to local agribusiness exchanges) Source: authors. Instrument: Diversification of agricultural sector by promoting non-traditional export products ARM Strategy: Mitigation Relevance to Paraguay: Soy and beef alone comprise over a third of Paraguay’s total exports. Given this high concentration of economic activity in two commodities, Paraguay is exposed to adverse shocks in terms of trade. Volatility in price (both of inputs and exports) and exchange and interest rate volatility increase revenue uncertainty. Paraguay could mitigate risk by diversifying its portfolio of exports. Other countries like Peru have employed different methods in the past to diversify production and support non-traditional exports. Peru’s economy has suffered from strong swings in terms of trade due to concentration of export products. Exports in Peru have historically concentrated on a few primary products, mainly fishmeal and minerals. Export diversification is a strategy to reduce market risk. Resource-rich, Peru has tried a number of policies to diversify the export base, moving from protectionist policies in the 1970s and 1980s towards liberal reforms in the 1990s.40 40 Illescas, Javier and Jaramillo (2011). 48 Non-traditional agricultural exports have increased and diversified significantly in the past decade. Peru is expanding exports of products that have not been previously exported as well as increased the export of products to new destination markets. From 2000 to 2005, the average annual rate of growth for non-traditional agricultural exports was 20 percent and reached a value of $1.02 billion in 2005.41 Asparagus has been one of the most successful non-traditional agricultural exports, with a 25 percent share of the total value of all non-traditional exports in 2005.42 Public and private sector cooperation in the development of the asparagus industry, coupled with favorable exogenous economic conditions and opportunities, contributed to investments in quality improvements, product safety, logistics efficiency, and coordination of actors along the supply chain. The Government lent support to the expansion of drip irrigation which was necessary for asparagus to take off. The Peruvian export promotion agency, helped establish a non-profit, the Peruvian Institute of Asparagus (IPE), which went on to negotiate for preferential US tariffs for Peruvian asparagus and develop integrated pest management and sanitary certifications. The public sector also helped with coordination issues between importers and exporters and improvements in logistics efficiency like a cold chain organized by Frio Aereo. 43 Today, Peru is the world’s leader in exports of green asparagus. There are many instruments for Governments to promote export diversification. Such measures include decoupled subsidies to promote diversification (in compliance with the WTO) and removing subsidies for traditional crops. Table 3. 7: Peru’s study case. Benefits, Challenges and considerations for Paraguay Benefits Challenges Considerations for Paraguay  Non-traditional  Coordination and clarity in roles between  Analyze current subsidies and exports can represent public and private sector actors to align export engagement to ensure a high-value niche incentives, knowledge transfer, and alignment with development market and be marketing. Identifying market failures objectives especially profitable (and Government failures) help determine  Select the appropriate new given first-mover roles for the public and private sector. products to fit Paraguay’s advantages.  Switching to non-traditional, high-value development goals, assets, and  Diversifying export products require complementary comparative advantage production can offset investments in logistics, technology, and  Define public sector volatility in the inputs. Access to credit and markets are engagement. Subsidies should markets for other also important factors for success. meet criteria for efficiency and products.  Niche markets are also subject to volatility be compliant with WTO in demand and supply and will evolve over regulations. time as other producing countries enter the  The ecological impact of market. expanding new crop should be  Equitable access to new agricultural considered (Peru is facing issues technology and crops can be difficult; not with water consumption of all farmers are able to switch due to capital asparagus production). Poorly constraints and lack of market integration. managed natural resources can increase risk in the medium to long term. 41 Rios (2007). 42 Ibid. 43 Shimizu (2006). 49 Source: authors. Instrument: Promotion of forward contracts and subsidies for premiums on options (derivative) contracts bought in international commodity exchanges ARM Strategy: Transfer Relevance for Paraguay: The three main agriculture commodities in Paraguay (soy, meat and maize) have liquid futures/options markets where coverage against price fluctuations can be bought. Currently, agro-exporters are (for the most part) the ones in Paraguay who purchase these price hedging contracts, as well as some large and integrated farmers (in particular those selling through the Brazilian market). However, the Government and small farmers are left retaining the risk of such commodity price volatility, undermining shared prosperity and their capacity to accumulate capital and smooth income. Agricultura por Contrato (AxC) is a price risk management program initiated in 2001 as part of a broader program called “Programa de Prevencion y Manejo de Riesgos�. This larger program included sub-programs to support the production of specific crops, the commercialization and export of specific crops, quality certification, access to grains for animal production, and contract farming for livestock producers. The two main subprograms are: i) establishment of fixed bases (differentials) over Chicago Board of Trade (CBOT) futures prices for each of the two main agricultural seasons in a given year and provision of compensation to producers and consumers when prices moved away from those fixed bases, hereafter referred to as the bases compensation program; and ii) co-financing of the purchase of options (puts and calls) used to hedge the physical forward contracts agreed between producers and consumers, hereafter referred to as the hedging program. Mechanically, the risk management programs provide compensation which protects participants from volatility in the physical price of key commodities for a select set of Government-supported forward contracts, called AxC contracts. This is designed to encourage producers and consumers to engage in more forward contracting, which in turn supports the commercialization of agricultural trade. Forward contracts for the central commodities covered by the program (yellow maize, wheat, sorghum) are typically priced by taking the CBOT futures prices and adding a premium.44 The premium is calculated as a differential (bases) that reflects local supply and demand conditions, and the costs of logistics, insurance, and financing. Producers and consumers agree on these contracts at the beginning of a production cycle (pre-harvest), but do not deliver and settle the contracts until the end of the production cycle three to six months later (post- harvest).45 In between, the bases component of the price can fluctuate, creating risk for both the 44 Cotton, coffee, orange juice, pork, beef, and white maize were added in 2010. 45 Throughout this chapter the word “consumer� refers to buyers of the physical commodity, generally local and international agribusinesses, trading companies, and market intermediaries. The word “final consumer� refers to 50 producer and the consumer. The diversity of Mexican agricultural production/consumption patterns, along with its size and geography, create high levels of differentiation between markets. This means that the number of bases differentials, which correspond to individual commodities and production/consumption zones, is high. The bases compensation program targets the estimated bases price levels, above the CBOT futures price, for each of the two main harvest seasons (spring/summer and winter/fall). Estimated bases levels for specific commodities produced in targeted states are announced at the beginning of the season, and are used to establish the pricing for the physical AxC contracts agreed between the producers and consumers. The bases levels are derived using a formula that starts with the CBOT futures price, and then adds the costs for physical delivery of maize from the US to a specific consumer zone in Mexico (referred to as the Standardized Base Consumer Zone). Costs for physical delivery from a local production zone to the consumer zone (referred to as the Standardized Base Production Zone) are then subtracted to determine a producer price. At the end of the season, actual bases price levels are calculated using the average of prices for physical transactions observed during the first fifteen days of the harvest (for each corresponding crop and production cycle) and the actual transportation costs observed during the same time window. The program then provides compensation to producers and consumers for negative movements between the estimated bases and the actual bases which may have occurred during the period between agreement (pre-harvest) and settlement (post-harvest) of the forward contract. When the actual calculation of the bases price is higher than the estimated level, payment goes to the producer, thereby ensuring that he/she is being compensated for the increase in prices reflected by the current market at harvest time. When the actual calculation of the bases price is lower than the estimated level, payment goes to the consumer thereby ensuring that he/she is able to take advantage of the decrease in prices reflected by the current market at harvest time. The hedging program targets the CBOT futures price which is used as the price reference for negotiation between producers and consumers of the AxC physical contract. Under the program previously managed by ASERCA, producers and consumers entering into AxC contracts were provided with option contracts which hedged the CBOT futures price level fixed in the AxC contract. The main objective of this approach was to reduce the incentive to default on the forward AxC contracts in the event of favorable price movements (up for producers, down for consumers) which could occur in between the time of contract agreement (pre-harvest) and contract settlement/delivery (post-harvest). The hedging program therefore provided producers with a call option (which would provide a payout if market prices increased) and consumers with a put option (which would provide a payout if market prices decreased). ASERCA took responsibility for purchase of the option contracts, which were also settled by ASERCA at the end of the season. Table 3.8 summarizes the levels of subsidy support provided to producers and consumers under the program as managed by ASERCA. It is important to differentiate between the two different types of subsides: for the bases compensation program, Government provides a cash payment which offsets movements in the bases price between pre- and post-harvest period; for the hedging program, Government provides a subsidy to cover a portion of the cost (premium) used to purchase a CBOT option contract. Recently the ASERCA program has been reactivated and now coexists with a newer program. individual who is a purchaser of a food product, typically purchasing that product at further points along in the supply chain. 51 Table 3. 8: Subsidy Components, AxC Program, ASERCA Component Producer Consumer Bases compensation program Differential between estimated Differential between estimated standardized bases and actual, if standardized bases and actual, if actual > estimated actual < estimated Hedging program 80 percent coverage for purchase of 70 percent coverage for purchase of a call option a put option Source: adapted from materials provided by ASERCA, August, 2012. Benefits Challenges Considerations for Paraguay  Income smoothing for  Very costly program, which is  Who would be the target both buyers and sellers of undergoing reforms due to budget beneficiaries of such program in agriculture commodities constraints. order to ensure the progressivity in Mexico  The program (as currently design) of the subsidy?  The effects of an increase does not promote the development of  How to promote the use of the in the use of forwards in a a local futures/options contract. price hedging instruments by market can change  Basis risk (price differential in prices farmers (education)? investment decisions by between CBOT and domestic  What to do about the price buyers and sellers of the markets). differential between Paraguay commodity.  Program eligibility is broad, and thus and the Brazilian, Argentinean most benefits go to larger and Chicago market prices? producers/buyers who may not need the support. Source: authors. Instrument: Soy futures contract at ROFEX, Rosario’s Agricultural Futures and Options Exchange ARM Strategy: Transfer Relevance for Paraguay: As in Argentina, Paraguay exports soy and most of soy exporters of Paraguay are a subsidiary or part of the same company that operates in Argentina. On the other hand, small farmers in Paraguay do not have access to such hedging instruments, while in Argentina they do through cooperatives operating in local exchanges. The Bolsa de Comercio de Rosario (BCR) (established in 1884), is a not-for-profit organization based in Rosario, Province of Santa Fe, Argentina. BCR is the main grain cash market of Latin America, larger than Kansas [KCBT] and second in rank to Minneapolis [MPLS] in the U.S. 46 It serves as a forum for the conduct of trade negotiations in several markets. It hosts four Divisions, Markets, or Exchanges: (i) the BCR itself as a huge cash market in several agricultural products and other services, like modern grain and oilseed laboratories; (ii) cámara Arbitral de Cereales de la Bolsa de Comercio de Rosario S.A.; (iii) mercado a Término de Rosario S.A. -Rosario Futures Exchange – ROFEX- [1909]; and (iv) mercado de Valores de 46 In 2010 KCBT delivered 775,200 tons; MPLS 234.2 million tons # 1 in tonnage delivered worldwide. 52 Rosario S.A. [The Rosario Securities Exchange]. Another recent market could be added (v) an independent virtual Cattle Market via TV weekly sessions. The BCR’s physical or grain cash-market is the most important in Argentina and in Latin America, in terms of weight and value, also providing reference prices for the main national and international grain markets. The BCR additionally operates a huge complex of laboratories through other of its affiliations, Cámara Arbitral de Cereales de Rosario S.A. [CACBCR], where thousands of samples are (anonymously) analyzed using code bars, providing certainty to clear disputes and quality certifications for all type of agricultural products, by- products, and soil and water analysis, as well. Based on a dynamic and well established cash market, ROFEX was founded in 1909 as the “Mercado General de Productos Nacionales� (General Exchange of National Commodities). In 1924 this market traded futures and forwards of linen (1.3 million tons); wheat (3.7 million tons); and corn (3.3 million tons in 1929). The market activity went on until trading was hampered by interventionist Governments (1932 recession, World War II, etc.) Government intervention and subsequent chronic inflation. ROFEX has traditionally been a futures exchange for commodities. Between the late 1930s and 1980s, the Exchange was used by Government as a regulatory agency and for official grain purchases. At the start of the nineties (1991-1993) the Government allowed the negotiation of live-cattle and grain futures and options contracts in US dollars, the second being a cash-settled soybeans contract, named “�ndice Soja Rosafé� or ISR,47 the main agricultural hedging instrument in Rosario, among other standardized delivery contracts (corn, wheat, soybeans, and sunflower) traded at the Buenos Aires Futures Exchange (Mercado a Término de Buenos Aires S.A., MATBA), located at the Buenos Cereal Exchange (Bolsa de Cereales de Buenos Aires, BCBA). Table 3.10 : Comparison CBOT, MATBA, & ROFEX CBOT MATBA ROFEX Tons 136 100 30 Margin U$s 3.375,- u$s 1.632,- u$s 1.632,- Fee u$s 1,90 0,5% s/MC u$s 12,20 Spread u$s 10,- u$s 20,- u$s20,- Source: authors. The success in launching a new futures/options contract on agricultural commodities is the value added to market participants. In the case of Argentina and ROFEX, this was not only based on the price differential between the Argentine Soy Market and CBOT, but also on the size of the CBOT contracts. ROFEX offers a smaller contract, which makes it possible for smaller market participants (farmers, cooperatives, intermediares, buyers) to access such price hedging products and in the local currency (Table 3.10). 47 By the end of 2010 the ISR represented more than 75 percent of all Rofex’s contracts. In 1991 and 1992 the ISR’s options on futures was the first of its kind in Argentina. 53 There are some pre-conditions that are important for Paraguay to consider in terms of having a local futures contract for soy or any other agriculture commodity (livestock, maize). The main ones include: i) market size, liquidity and minimum volume of contracts: a reduced number of agribusinesses (i.e. monopsonies) reduces the benefit from trading using standard contracts or from making the price of financial or physical products public. There need to be sufficient contracts for physical commodities (or financial products) so that there is competition for buying, selling, and keeping long or short open positions, so that price volatility is reduced due to the law of large numbers; ii) institutional and regulatory framework: A set of clear and transparent rules for market participants and potential entrants is important. The lack of transparent rules for entrants, for instance, has stifled competition in some countries. Regulations about membership procedures and limits, or minimum capital and trade requirements, should be negotiated by participants and published. Furthermore, exchanges may attract both local and international capital. Consequently, financial regulations will be required up to international standards. Sound financial management is also required, since international investors may demand dollar or euro-based contracts; iii) committed agribusinesses: agribusinesses are essential for the development of agricultural futures contracts, as they are the main beneficiaries. Supermarkets, processing and trading plants, food exporters/importers, agricultural input suppliers, and farmers must see the benefit of such contracts. Some of the elements that agribusinesses consider for deciding whether to support or not trading through the exchange are (the list is not exhaustive): (i) tax benefits; (ii) benefits from having a locally publicly known reference price; (iii) benefits from a third party quality control mechanism; (iv) in some cases the quality standards, controls and grades agreed to be traded through the exchange; and (v) arbitration mechanism. Another aspect for agribusiness is the value added of the contract versus contracts already being traded in other exchanges nearby. There needs to be a differentiation of contracts to avoid competition and to lower basis risk. 48 Because international agricultural commodity exchanges already carry a wide variety of standard agriculture commodity contracts, a futures contract in Paraguay will struggle if it replicates an already existing contract, in particular for commodities whose prices are highly correlated with international markets. Table 3.11: Argentina’s study case. Benefits, challenges and considerations for Paraguay Benefits Challenges Considerations for Paraguay  Increase access to agriculture price  Achieve enough liquidity to  Examine price differentials hedging instruments by all market sustain the local between domestic and participants (in particular smaller futures/option international commodity buyers and sellers).  Strengthen the commodity markets.  Reduce the cost and basis risk of exchange and the regulator to  Discuss value added of new commodity price hedging for all operate and supervise the contracts with agribusinesses market participants, in particular in market. and agroexporters. relation to exchange rate risk and  Educate the potential market  Identify what commodities price differentials. participants in the usage and could be initially traded in  Promoting local agriculture price operation of futures/options order to create success stories formation contract. to gain momentum. Source: authors. 48 Basis risk here is defined by the difference in price movements between the international market (i.e. CBOT) and the domestic market that would trade such commodity. If the basis risk is low, agribusinesses would not see the need to develop a local commodity exchange, and would thus use the international market to trade/hedge. In many cases exporters and arbitrageurs would go long buying cash, forward or futures contracts in one market and, simultaneously or at a later time, would sell or go short in the delivery market. 54 3.3 A comprehensive approach to managing volatility It is important to develop a comprehensive macroeconomic risk management framework that takes all different sources of volatility and risks into account. Different sources of volatility do not occur in isolation; the analysis in this study has shown that there are many interrelations between different variables; some may lead others, or there may be self-reinforcing mechanisms at play as for example in the case of investment and growth volatility. In addition, policy options available to address these different sources of volatility also feed back into the system, partly in the desired way and partly with side effects. For example, what would be the budget implications for Paraguay if the Government were to introduce a contingency fund for small farmers similar to the Mexican CADENA program? How big would the fund have to be to effective and to have a distributional impact? How big would be potential contingent liabilities and how would this impact fiscal sustainability? What would be the distributional impact of an introduction and how would that fit with the overall design of the social protection system? These interrelations and examples illustrate the need for a careful assessment of choices and emphasize the need for a comprehensive macroeconomic risk management framework to develop priorities under the constraints of overall macroeconomic management. Taking a broader perspective is recommended because it allows for finding optimal ways to manage the observed volatility and risks. For instance, looking narrowly at ways of managing commodity price volatility one would find a set of useful tools, and yet, this approach would not include considering a set of policy options which could be effective in the medium- and long-run and could at the same time also alleviate additional challenges that the Government in Paraguay is facing: Government taxation itself contributes to a situation in which agriculture weighs heavy in Paraguayan GDP and exposes the country to the fluctuations of commodity prices. The fact that the agricultural sector is practically exempt from taxation, despite being one of the most important corporate sectors of the country, introduces a distortion into the economy which sets disincentives for entrepreneurs to explore business opportunities in other sectors. Moreover the low taxation of agriculture contributes to the low tax-to-GDP ratio in Paraguay, which is a major challenge to the public provision of social services. Certainly, addressing tax distortions is complementary to looking for more targeted ways to manage a specific type of volatility. And the strategy of assessing possibilities to diversify the economy is also linked back to specific tools of agricultural risk management as illustrated by the Peruvian case study and the development of asparagus market. The policy options presented in this study are intended as a basis for a dialogue on possible steps towards further improved management of growth volatility in Paraguay. The analytical background, which provides some information on the sources and the effects of growth volatility in Paraguay is intended as a starting point for further research on the topic and a dialogue on possible policy options. The options presented in this chapter need to be carefully assessed from the vantage point of applicability to Paraguay. They are an invitation for discussion and further investigation. Questions whether the institutional framework in Paraguay is adequate for hosting a fiscal rule or a stabilization fund, or whether weather data has sufficient quality to allow for the creation of weather related derivatives need to be assessed. The Government and the World Bank have been engaging in a dialogue on this topic through the preparation of this study and with a joint agricultural risk management assessment. 55 Acosta-Ormaechea, S. (2011): Paraguay—Potential output growth and spillovers from agriculture�, Selected Issues IMF country report No. 11/239, August. Alouini O. and P. Hubert (2010): “Country size, Growth and Volatility“, Documents de Travail de l'OFCE 2010-18, Observatoire Francais des Conjonctures Economiques (OFCE). Andrews, D. W. K. (1993): “Tests for Parameter Instability and Structural Change with Unknown Change Point�, Econometrica, 61, 821-856 Athanasoulis, S., and E. van Wincoop (2000): “Growth Uncertainty and Risk-Sharing.� Journal of Monetary Economics 45(3):477–505. Berument, H. (2013): “Agricultural Performance and Macroeconomic Outcomes in Paraguay�, background paper for World Bank report: �Growth volatility in Paraguay—sources, effects, options�, World Bank 2013, (forthcoming). Berument, H. (2013a): “Paraguayan Agricultural and macroeconomic performance: a wavelet approach�, background paper for World Bank report: �Growth volatility in Paraguay— sources, effects, options�, World Bank 2013, (forthcoming). Berument, H., Dincer, N. and Z. Mustafaoglu (2011): “Total factor productivity and macroeconomic instability�, The Journal of International Trade & Economic Development, Vol. 20, No. 5, pp. 605–629. Borda, D., Anichini, F. and J. Ramirez (2013): “Study of Agricultural Volatility in Paraguay“, Volume 2 of World Bank report:�Growth volatiltity in Paraguay—sources, effects, options�, World Bank 2013, Washington D.C., (forthcoming). Breen, R. and C. García-Peñalosa, 2005. "Income Inequality and Macroeconomic Volatility: An Empirical Investigation,"Review of Development Economics, Wiley Blackwell, vol. 9(3), pages 380-398. Byrne, J. P., Fazio, G., and N. Fiess (2012): “Primary commodity prices: co-movements, common factors and fundamentals�, Journal of Development Economics, 101, 16-26 Calderon, C. and K. Schmidt-Hebbel (2003): "Macroeconomic policies and performance in Latin America,"Journal of International Money and Finance, Elsevier, vol. 22(7), pages 895- 923, December. Calvo, G. (2008): “Exporting Commodity Prices, Lax Monetary Policy and Sovereign Wealth Funds�, Vox EU, June 20. http://www.voxeu.org/article/exploding-commodity-prices- signal-future-inflation. 56 Calvo, G., Leiderman, L. and Reinhart, C. (1993): "Capital Inflows and Real Exchange Rate Appreciation in Latin America: The Role of External Factors", IMF Staff Papers, Palgrave Macmillan, vol. 40(1), pages 108-151. Calvo-Gonzalez, Oscar, Rashmi Shankar, and Riccardo Trezzi (2010): "Are Commodity Prices More Volatile Now." A Long-Run Perspective‖ , World Bank Policy Research Working Paper 5460. Consolidación de la política sanitaria y de inocuidad para las cadenas lactea y cárnica. Conpes, Jornal de Política Económica y Social de Colombia. Departamento Nacional de Planeación. 19 Julio 2010, available at: http://www.minagricultura.gov.co/archivos/conpes_3676_politica_sanitaria.pdf. Cubas, L. Escobar, N. Franci, T. Olmedo, L. and Smith, P. (2011): “Época de Stroessner: economía paraguaya en la dictadura (1954-1989)�, Mimeo, Universidad Nacional de Asunción. Cuesta, J. and P. Suárez Becerra (2013). Equality of Opportunities and Public Spending in Paraguay. Background paper prepared for the Paraguay PER. Dana, J. (2012): “Market-based approaches to managing food security risks: a review of instruments and applications for Governments.� Davis, J., Ossowski, R., Daniel, J., and S. Barnett (2001) “Stabilization and Savings Funds for Nonrenewable Resources – Experience and Fiscal Policy Implications,� International Monetary Fund. Debrun X., D. Hauner, and M. Kumar (2009), “Independent Fiscal Agencies,� Journal of Economic Surveys, Vol.23, No. 1, pp.44-81. Diaz Bonilla, C. and Cicowiez, M. (2013):�Assessing the Poverty and Social Impact of Tax Exemptions and World Price Shocks in Paraguay�, Volume 2, Paraguay Public Expenditure Review, The World Bank, Washington D.C.. Favaro, E., Koehler-Geib, F., Picarelli, N., and A. Indaco (2013): Paraguay: Agriculture commodity prices and tax revenue collection, background paper for Paraguay Public Expenditure Review (2013), World Bank, forthcoming Frankel, J. (2008): “The Effect of Monetary Policy on Real Commodity Prices�, in Campbell, J.Y. (Ed.), Asset Prices and Monetary Policy, NBER, University of Chicago, Chicago, pp. 291-333. 57 Frankel, J and A. Rose (2010): “Determinants of Agricultural and Mineral Commodity Prices�, HKS Faculty Research Working Paper Series RWP10-038, John F. Kennedy School of Government, Harvard University. Frankel, J., Vegh, C, and G. Vuletin (2013): “On graduation from fiscal procyclicality�, Journal of Development Economics, Elsevier, vol. 100(1), pages 32-47. Furceri, D. and Karras, G. (2007): "Country size and business cycle volatility: Scale really matters," Journal of the Japanese and International Economies, Elsevier, vol. 21(4), pages 424-434, December. Gausch, J. (2010): Export Diversification and Competitiveness: How to Grow and Alleviate Poverty Through Exports, World Bank, Washington, D.C. Garcia-Penalosa, C. and Turnovsky, S. (2004): “Macroeconomic Volatility and Income Inequality in a Stochastically Growing Economy�, conference volume of “Economic Growth and Income Distribution: On the Nature and Causes of the Wealth of Nations,� conference held in Lucca, Italy, June 16-18. Gavin, M., Hausmann, R., and L. Leiderman (1995): "Macroeconomics of Capital Flows to Latin America: Experience and Policy Issues", Research Department Publications 4012, Inter- American Development Bank, Research Department. Goñi, E., J.H. López and L. Servén (2008). Fiscal Redistribution and Income Inequality in Latin America. Policy Research Working Paper No. 4487. Washington, DC: World Bank. Hnatkovska, V. and F. Koehler-Geib (2013):“Business Cycle Accounting for Paraguay“, Volume 2 of World Bank report:�Growth volatiltity in Paraguay—sources, effects, options�, World Bank 2013, Washington D.C., (forthcoming). Hnatkovska, V., and N. Loayza (2003): “Volatility and growth�. World Bank Policy Research Working Paper No. 3184, World Bank, Washington, D.C. Huang, H., Fang, W. and S. Miller (2012): “The Effect of Growth Volatility on Income Inequality�, University of Nevada Working paper. Illescas, Javier and C. Felipe Jaramillo (2011). Export Growth and Diversification: The case of Peru. Policy Research Working Paper, World Bank, Washington D.C. Ilzetzki, Ethan and Carlos A. Vegh, 2008. "Procyclical Fiscal Policy in Developing Countries: Truth or Fiction?," NBER Working Papers 14191, National Bureau of Economic Research, Inc. IMF (2012):“Strengthening Government debt and cash management�, IMF report, International Monetary Fund, Washington, D.C., March 2012. IMF (2011):�Paraguay Article IV consultations—staff report�, International Monetary Fund, Washington, D.C., August 2011. 58 IMF (2009): “Paraguay, addressing the stagnation and instability trap�, edited by Alejandro Santos, International Monetary Fund, Washington DC. IMF (2009a) “Fiscal Rules – Anchoring Expectations for Sustainable Public Finances,� International Monetary Fund, Washington, D.C. Inclan, C. and G. Tiao (1994):“Use of Cumulative Sums of Squares for Retrospective Detection of Changes of Variance,� Iturrioz, R. and D. Arias (2011), Agricultural Insurance in Latin America: Developing the Market, World Bank, Washington D.C. Kokoszka, P. Leipus,R. (2000): “Change-point estimation in ARCH models�, Bernoulli, 6: 513- 539. Kose, M. A., Otrok, C., and E. Prasad (2012): “Global Business Cycles: Convergence Or Decoupling? “, International Economic Review, Department of Economics, University of Pennsylvania and Osaka University Institute of Social and Economic Research Association, vol. 53(2), pages 511-538, 05. Kose, M. A., Otrok, C., and C. H Whiteman (2008): “Understanding the evolution of world business cycles“, Journal of International Economics, Elsevier, vol. 75(1), pages 110-130, May. Le Fort, G., Escobar, G., and D. Contreras (2013):�Paraguay: Estimation of the Structural Fiscal Balance and Fiscal Rule Proposal�, Volume 2, Paraguay Public Expenditure Review, The World Bank, Washington D.C.. Lopez-Calva, L., Lugo, M. and O. Barriga Cabanillas (2013):�Paraguay Equity Assessment�, The World Bank, Washington, D.C., forthcoming. Loayza, Ranciere, Serven, and Ventura (2007):� Macroeconomic Volatility and Welfare in Developing Countries: An Introduction�, The World Bank Economic Review, Vol. 21, No. 3, pp. 343–357. Lustig, Nora, Sean Higgins, Julio Ramirez and Billy Swanson (2012). Social Spending, Taxes and Income Redistribution in Paraguay. Background paper prepared for the Paraguay PER. Mahul, O. and J. Skees (2006): “Piloting Index-Based Livestock Insurance in Mongolia�. Mahul, O., N. Belete, and A. Goodland (2009): “Innovations in Insuring the Poor: Index-based Livestock Insurance in Mongolia�, Focus 17, Brief 9, December 2009, available at: http://www.ifpri.org/sites/default/files/publications/focus17_09.pdf. 59 Mendoza, Enrique G. and P. Marcelo Oviedo, 2006. "Fiscal Policy and Macroeconomic Uncertainty in Developing Countries: The Tale of the Tormented Insurer," NBER Working Papers 12586, National Bureau of Economic Research, Inc. Nerlove, M. (1958): “The Dynamics of Supply: Estimation of Farmers' Response to Price�, Baltimore: The Johns Hopkins Press. Neumeyer, P. and F. Perri, (2005): "Business cycles in emerging economies: the role of interest rates," Journal of Monetary Economics, Elsevier, vol. 52(2), pages 345-380, March. OECD (2009): “OECD Economic Surveys: Mexico,� OECD, Paris. Paes de Barros, R., F.H.G. Ferreira, J.R. Molinas Vega and J. Saavedra Chanduvi (2009): “Measuring Inequality of Opportunities in Latin America,� The World Bank, Washington, DC. Paz et al. (forthcoming) for Bolivia, Higgins and Pereira (forthcoming) for Brazil, Scott (forthcoming) for Mexico, Jaramillo (forthcoming) for Brazil Perry, G. and N. Fiess (2006): “Turmoil in Latin America and the Caribbean: Volatility, Spillovers, and Contagion�, World Bank working papers book 3. Podpiera, J. and V. Tulin (2012): “External financial shocks: How important for Paraguay?�, Selected Issues IMF country report No. 12/212, August. Raddatz, C. (2007): “Are external shocks responsible for the instability of output in low-income countries?, “ Journal of Development Economics, Elsevier, vol. 84(1), pages 155-187, September. Rios, L. (2007): “Agro-industries characterization and appraisal: Asparagus in Peru. Agricultural Management,� Marketing, and Finance Working Document, FAO, available at: http://www.fao.org/docrep/016/ap297e/ap297e.pdf. Schmidt-Hebbel, K. (2012) “Fiscal Policy for Commodity Exporting Countries: Chile’s Experience,� Documento de Trabajo No. 415, Pontificia Universidad Catolica De Chile, Instituto de Economia. Shimizu, Tatsuya (2006). Expansion of Asparagus Production and Exports in Peru. Institute of Developing Economies Discussion Paper, available at: http://ir.ide.go.jp/dspace/bitstream/2344/145/1/ARRIDE_Discussion_No.73_shimizu.pdf. Syroka, Joanna (2007). Experiences in Index-Based Weather Insurance for Farmers: Lessons Learnt from India and Malawi. Commodity Risk Management, Agriculture and Rural Development, World Bank. Presentation to the Innovative Finance Meeting, New York, available at: http://www.agroinsurance.com/files//syroka.pdf. 60 Talvi, E., and C. Végh, 2005. “Tax Base Variability and Procyclical Fiscal Policy in Developing Countries,� Journal of Development Economics 78, pages 156-190. United Nations Development Program (2006): “Usos y gobernabilidad del agua en Paraguay�, http://www.undp.org/content/dam/paraguay/docs/usoygobernabilidad.pdf Uribe, M. and V. Yue (2006): "Country spreads and emerging countries: Who drives whom?," Journal of International Economics, Elsevier, vol. 69(1), pages 6-36, June. USDA (2012) Commodity Intelligence Report, Foreign Agricultural Service, available at: http://www.pecad.fas.usda.gov/highlights/2012/02/Paraguay_Drought/. World Bank (2014):�Managing Risk for Development�, World Development Report 2014, the World Bank, Washington D.C., forthcoming. World Bank (2013):�A Public Expenditure Review for Paraguay—The Quest for Optimal Tax and Expenditures Policies�, The World Bank, Washington D.C., forthcoming. World Bank (2013a): “Improving Resilience and Reducing Vulnerability--Agricultural Sector Risk Management –Technical Assistance from the World Bank's Agricultural Risk Management Team�, The World Bank, Washington D.C., available at http://go.worldbank.org/3VZO2PE7Q0. World Bank (2013b): “Analisis de reisgo del sector agropecuario�, The World Bank, Washington D.C., available at http://go.worldbank.org/3VZO2PE7Q0. World Bank (2012):�Public Expenditure Review for Peru—Spending for Results�, The World Bank, Washington D.C. World Bank (2011): “Weather Index Insurance for Agriculture: Guidelines for Development Practitioners,� Agriculture and Rural Development Discussion Paper 50, World Bank, Washington D.C., available at: https://www.agriskmanagementforum.org/sites/agriskmanagementforum.org/files/WEAT HER%20INDEX%20INSURANCE%20FOR%20AGRICULTURE_WB_WII_Paper_No v_2011.pdf. World Bank Treasury (2011): “Mitigating the Impact of Drought on Food Security in Malawi— Case Study, World Bank, Washington D.C., available at: http://treasury.worldbank.org/bdm/pdf/Case_Study/Malawi_WeatherHedge_FoodSecurit y.pdf. World Bank (2000):�Securing Our Future in a Global Economy�, The World Bank, Washington, D.C. World Bank and IMF (2012) Joint WB-IMF MTDS report, October, 2012. 61 World Bank and IMF (2010):“Paraguay-Compendium of recommendations on debt management� Joint WB-IMF MTDS report, December, 2010. Quandt, R. E. (1960): “Tests of the hypothesis that a linear regression system obeys two separate regimes,� Journal of the American Statistical Association, 55, 324-330 Zhu (2003). Quantitative models for performance evaluation and benchmarking: data envelopment analysis with spreadsheets and DEA excel solver. Massachusetts: Kluwer Academic Publishers. 62 Annex 1.1: Volatility over time, international comparison Standard deviation of Paraguay’s GDP growth and output gap in international comparison std dev (GDP growth) std dev (GDP gap) 1960- 1960- 2000- 1960- 1960- 2000- 2011 1999 2011 2011 1999 2011 Argentina 5.83 5.56 6.73 5.66 5.20 6.86 Bahamas, The 7.16 7.91 2.62 7.87 8.84 2.95 Barbados 4.54 4.61 3.51 4.46 4.69 3.62 Belize 4.03 4.17 3.56 4.81 5.30 2.60 Bolivia 3.52 3.93 1.31 3.98 4.49 1.37 Brazil 4.11 4.51 2.29 3.84 4.30 1.63 Chile 4.64 5.21 2.02 4.50 5.05 1.73 Colombia 2.21 2.35 1.77 2.31 2.21 2.56 Costa Rica 3.34 3.49 2.85 3.32 3.49 2.77 Cuba 6.36 7.02 3.65 6.52 7.27 3.90 Dominican Republic 5.26 5.75 3.36 4.63 4.98 3.41 Ecuador 3.55 3.82 2.48 3.17 3.25 2.94 El Salvador 4.18 4.66 1.84 4.63 5.20 1.89 Guatemala 2.49 2.73 1.46 2.56 2.83 1.40 Guyana 5.22 5.74 2.84 5.18 5.69 2.61 Honduras 3.04 3.24 2.42 3.09 3.17 2.92 Jamaica 5.03 5.18 0.33 5.20 5.31 0.26 Mexico 3.78 3.78 3.34 3.25 3.39 2.82 Nicaragua 6.23 7.06 1.96 5.70 6.41 2.12 Panama 4.40 4.56 3.67 4.14 4.34 3.43 Paraguay 4.28 3.88 5.50 4.31 4.22 4.45 Peru 5.03 5.39 3.14 5.01 5.53 2.69 Puerto Rico 3.55 3.10 2.78 2.79 2.73 3.06 Suriname 5.24 5.69 2.10 4.50 5.15 2.68 Trinidad and Tobago 4.99 4.70 5.71 5.36 4.98 6.62 Uruguay 4.44 4.26 5.12 5.37 5.28 5.53 Venezuela, RB 5.32 4.36 7.90 5.17 3.90 8.24 LAC mean 4.52 4.72 3.11 4.50 4.73 3.18 (excluding Paraguay) LAC median 4.49 4.59 2.81 4.57 4.98 2.80 (excluding Paraguay) Mercosur 3.53 3.68 3.10 3.37 3.47 2.99 (excluding Paraguay) East Asia & Pacific 2.77 2.90 1.83 2.01 2.12 1.64 (all income levels) Europe & Central Asia 1.89 1.71 2.22 1.61 1.47 2.04 (all income levels) 63 Middle East & North 3.78 4.35 1.65 3.11 3.53 1.64 Africa (all income levels) South Asia 2.62 2.58 2.05 1.89 1.93 1.79 Sub-Saharan Africa 2.13 2.21 1.37 1.71 1.76 1.59 (all income levels) Lower middle income 1.66 1.58 1.50 1.45 1.44 1.54 Upper middle income 2.12 1.95 2.22 1.77 1.63 2.24 OECD members 2.01 1.77 2.06 1.48 1.37 1.85 Source: World Development Indicators, and Central Bank of Paraguay. Annex 1.2: Volatility breaks of macroeconomic variables in Paraguay Direction Standard Deviation Mean Coefficient Variable Date of change of variation Entire Period Before Break After Break* Entire Period Before Break After Break* in volatility GDP single breakpoint 2008-IV Increase 1.8 4.8 0.0 4.0 0.0 7.0 0.0 2.7 0.0 2.6 0.0 3.2 0.0 Agriculture sector single breakpoint 2008-IV Increase 3.1 11.9 0.0 6.3 0.0 22.5 0.0 3.9 0.0 4.9 0.0 0.4 0.0 Non -agriculture sector No change No change 1.6 4.2 0.0 -- 0.0 -- 0.0 2.5 0.0 -- 0.0 -- 0.0 Total investment No change No change 6.3 14.5 0.0 -- 0.0 -- 0.0 2.3 0.0 -- 0.0 -- 0.0 Private investment No change No change 6.6 22.2 0.0 -- 0.0 -- 0.0 3.4 0.0 -- 0.0 -- 0.0 Private consumption No change No change 1.8 5.4 0.0 -- 0.0 -- 0.0 3.0 0.0 -- 0.0 -- 0.0 Inflation single breakpoint 1995-II Increase 0.5 4.6 0.0 3.2 0.0 3.7 0.0 8.9 0.0 18.7 0.0 8.0 0.0 Soy price single breakpoint 2003-III Increase 3.2 27.1 0.0 18.2 0.0 31.6 0.0 8.5 0.0 0.9 0.0 14.8 0.0 Oil prices No change No change 2.1 36.8 0.0 -- 0.0 -- 0.0 17.5 0.0 -- 0.0 -- 0.0 Beef price No change No change 2.3 14.9 0.0 -- 0.0 -- 0.0 6.3 0.0 -- 0.0 -- 0.0 first breakpoint 2001-III Increase 7.2 8.6 11.5 35.1 Nominal Exchange rate second breakpoint 2003-II Decrease 2.1 14.5 6.2 6.9 -5.9 third breakpoint 2008-I Increase 0.0 0.0 12.6 0.0 0.0 0.0 -1.4 0.0 RER No change No change 8.3 10.5 0.0 -- 0.0 -- 0.0 1.3 0.0 -- 0.0 -- 0.0 TOT No change No change 26.1 16.0 0.0 -- 0.0 -- 0.0 0.6 0.0 -- 0.0 -- 0.0 Current account balance single breakpoint 2007-I Increase 3.4 146.6 0.0 97.0 0.0 202.9 0.0 43.3 0.0 18.4 0.0 74.7 0.0 World real interest rate single breakpoint 2007-IV Increase 2.2 2.4 0.0 1.9 0.0 2.4 0.0 1.1 0.0 1.8 0.0 -1.3 0.0 Interest rate No change No change 0.7 7.9 0.0 -- 0.0 -- 0.0 12.0 0.0 -- 0.0 -- 0.0 first breakpoint 2002-IV Decrease 11.4 8.1 12.3 -16.6 Credit to private sector 1.1 17.4 16.5 second breakpoint 2004-II Increase 0.0 0.0 14.2 0.0 0.0 0.0 26.3 0.0 Total revenue No change No change 0.7 9.6 0.0 -- 0.0 -- 0.0 13.2 0.0 -- 0.0 -- 0.0 Tax revenue single breakpoint 2004-III Decrease 1.0 15.0 0.0 21.2 0.0 8.7 0.0 14.4 0.0 14.6 0.0 14.2 0.0 first breakpoint 2000-II Increase 3.1 6.2 0.1 1.7 Public consumption 2.1 7.5 3.7 second breakpoint 2009-I Increase 0.0 0.0 6.9 0.0 0.0 0.0 13.6 0.0 first breakpoint 2002-II Decrease 55.6 17.0 14.9 5.1 Public investment 3.3 38.4 second breakpoint 2008-II Increase 38.3 11.5 16.6 Source: Central Bank of Paraguay. 64 65 Annex 1.3: Graphs on volatility breakpoints Inclan Tiao (1994) by variable Beef price and volatility break-points Current account balance and volatility break- points Stddev. of current account balance 35 16 250 400 14 25 300 200 12 Percent gropwth rate Stddev. of Beef 15 200 USD million 10 150 5 8 100 -5 0 100 6 -15 4 -100 50 -25 2 -200 -35 0 -300 0 Jul/ 97 Jul/ 98 Jul/ 99 Jul/ 00 Jul/ 01 Jul/ 02 Jul/ 03 Jul/ 04 Jul/ 05 Jul/ 06 Jul/ 07 Jul/ 08 Jul/ 09 Jul/ 10 Jul/ 11 Jul/ 12 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Beef price breaks stddev. Current account Balance breaks stddev. Sample statistics for entire series: Mean=6.3; Standarddeviation=14.9 Sample statistics for entire series: Mean=43.3; Standarddeviation=146.6 Private consumption and volatility break-points Public consumption and volatility break-points Stddev. of oil private consumption 6 8 Stddev. of oil public consumption 13 23 7 5 18 Percent growth rate 6 Percent growth rate 8 4 13 5 3 8 4 3 2 3 3 -2 -2 2 1 -7 1 -7 0 -12 0 Jan/ 95 Jan/ 96 Jan/ 97 Jan/ 98 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 Jan/ 95 Jan/ 96 Jan/ 97 Jan/ 98 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 Private consumption breaks stddev. Public consumption breaks stddev. Sample statistics for entire series: Mean=3.0; Standarddeviation=5.4 Sample statistics for entire series: Mean=3.7; Standarddeviation=7.5 Source: authors. 66 Interest rate Percent growth rate Percent growth rate -8 -3 -7 12 13 23 33 43 53 2 7 3 -27 -17 10 15 20 25 30 0 5 Jan/ 94 Jan/ 95 Jan/ 95 Jan/ 95 Jan/ 96 Jan/ 96 Jan/ 96 Jan/ 97 Jan/ 97 Jan/ 97 Jan/ 98 Jan/ 98 GDP Jan/ 98 Jan/ 99 Jan/ 99 Credit Jan/ 99 Jan/ 00 Jan/ 00 Interest rate Jan/ 00 Jan/ 01 Jan/ 01 Jan/ 01 Jan/ 02 Jan/ 02 Jan/ 02 Jan/ 03 Jan/ 03 Jan/ 03 Jan/ 04 Jan/ 04 Jan/ 04 points Jan/ 05 Jan/ 05 points Jan/ 05 breaks Jan/ 06 Jan/ 06 breaks Jan/ 06 Jan/ 07 Jan/ 07 breaks Jan/ 07 Jan/ 08 Jan/ 08 Jan/ 08 Jan/ 09 Jan/ 09 Jan/ 09 Jan/ 10 Jan/ 10 Jan/ 10 Jan/ 11 Jan/ 11 Jan/ 11 Jan/ 12 Jan/ 12 GDP and volatility break-point Jan/ 12 Domestic Interest rate and volatility break- stddev. stddev. Sample statistics for entire series: Mean=2.7; Standarddeviation=4.8 0 1 2 3 4 5 6 7 8 0 2 4 6 8 Credit to private sector and volatility break- Sample statistics for entire series: Mean=12.0; Standarddeviation=7.9 10 12 14 16 0 1 2 3 4 5 6 7 8 9 Sample statistics for entire series: Mean=16.5; Standarddeviation=17.4 stddev. St ddev. of GDP Stddev. of credit to private sector Stddev. of interest rate 67 Percent growth rate Percent Interest rate -4 -2 10 30 50 12 17 22 2 7 0 2 4 6 -50 -30 -10 Jan/ 98 Jan/ 94 Jan/ 94 Jan/ 99 Jan/ 95 Jan/ 95 Jan/ 96 Jan/ 96 Jan/ 00 Jan/ 97 Jan/ 97 Jan/ 01 Jan/ 98 Jan/ 98 Inflation Jan/ 02 Jan/ 99 Jan/ 99 Jan/ 03 Jan/ 00 Jan/ 00 Jan/ 01 Private investment Jan/ 04 Jan/ 01 Jan/ 02 International interest rate Jan/ 05 Jan/ 02 Jan/ 03 Jan/ 03 Jan/ 06 Jan/ 04 Jan/ 05 Jan/ 04 break-points Jan/ 07 Jan/ 05 Jan/ 06 breaks Jan/ 08 Jan/ 07 Jan/ 06 Jan/ 09 Jan/ 08 Jan/ 07 breaks Jan/ 09 Jan/ 08 Jan/ 10 breaks Jan/ 10 Jan/ 09 Jan/ 11 Jan/ 10 Jan/ 11 International interest rate and volatility Jan/ 12 Jan/ 12 Jan/ 11 Inflation and volatility break-points 0 5 3 3 3 3 3 3 4 4 4 0 1 1 2 2 3 3 Sample statistics for entire series: Mean=8.9; Standarddeviation=4.6 Sample statistics for entire series: Mean=1.1; Standarddeviation=2.4 10 15 20 25 stddev. Sample statistics for entire series: Mean=3.4; Standarddeviation=22.2 Private investment and volatility break-points stddev. Stddev. of private investment Stddev. of Inflation Stddev. of GDP stddev. Public consumption and volatility break-points Total investment and volatility break-points 130 60 16 Stddev. of oil public consumption 110 30 14 50 90 20 Percernt grosth rate Percent growth rate 12 70 Stddev. of Investment 40 50 10 10 30 30 8 0 10 20 6 -10 -10 4 -30 10 -20 -50 2 -70 0 -30 0 Jan/ 98 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 Jan/ 95 Jan/ 96 Jan/ 97 Jan/ 98 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 Public investment breaks stddev. Total investment breaks stddev. Sample statistics for entire series: Mean=3.7; Standarddeviation=7.5 Sample statistics for entire series: Mean=2.3; Standarddeviation=14.5 Nominal exchange rate and volatility break- Non-agriculture GDP and volatility break- points points 14.00 5 47 12 4 12.00 Stddev. of nominal exchange rate 37 Stddev. of non-agriculture GDP 4 Percent growth rate 10.00 Percent growth rate 27 7 3 17 8.00 3 6.00 2 2 7 2 (3) 4.00 -3 1 (13) 2.00 1 (23) 0.00 -8 0 Jan/ 95 Jan/ 96 Jan/ 97 Jan/ 98 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 Jan-92 Jan-93 Jan-94 Jan-95 Jan-96 Jan-97 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Nominal exchange rate breaks stddev. Non- agriculture breaks stddev. Sample statistics for entire series: Mean=6.9; Standarddeviation=14.5 Sample statistics for entire series: Mean=2.5; Standarddeviation=4.2 Oil prices and volatility break-points RER and volatility break-points 145 40 26 12 125 35 21 10 105 16 Percent growth rate 30 Percent growth rate Stddev. of RER Stddev. of oil prices 85 8 11 65 25 6 45 20 6 1 25 15 -4 4 5 10 -9 -15 2 -35 5 -14 -55 0 -19 0 Jan-98 Jan-99 Jan-00 Jan-01 Jan-02 Jan-03 Jan-04 Jan-05 Jan-06 Jan-07 Jan-08 Jan-09 Jan-10 Jan-11 Jan-12 Jan/ 96 Jan/ 97 Jan/ 98 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 Oil prices breaks stddev. RER breaks stddev. Sample statistics for entire series: Mean=17.5; Standarddeviation=36.8 Sample statistics for entire series: Mean=1.4; Standarddeviation=10.5 68 Fiscal revenue and volatility break-points Soy price and volatility break-points 25 35 53 43 63 30 20 Percent growth rate 33 25 Percent growth rate Stddev. of Soy price Stddev. of Revenue 43 23 15 20 23 13 15 10 3 3 10 -7 5 -17 5 -17 -37 0 -27 0 Jan/ 96 Jan/ 97 Jan/ 98 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 Soy price breaks stddev. Fiscal revenue breaks stddev. Sample statistics for entire series: Mean=14.4; Standarddeviation=15.0 Sample statistics for entire series: Mean=8.5; Standarddeviation=27.1 TOT and volatility break-points Revenue and volatility break-points 18 50 12 24 16 40 10 14 Percent growth rate 14 Percent growth rate Stddev. of TOT Stddev. of Revenue 12 30 8 4 10 20 6 -6 8 6 10 4 -16 4 -26 0 2 2 -36 0 -10 0 Jan/ 95 Jan/ 96 Jan/ 97 Jan/ 98 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 Jan/ 99 Jan/ 00 Jan/ 01 Jan/ 02 Jan/ 03 Jan/ 04 Jan/ 05 Jan/ 06 Jan/ 07 Jan/ 08 Jan/ 09 Jan/ 10 Jan/ 11 Jan/ 12 Total revenues of government TOT breaks stddev. breaks stddev. Sample statistics for entire series: Mean=0.6; Standarddeviation=16.0 Sample statistics for entire series: Mean=13.2; Standarddeviation=9.6 Source: authors. 69 Annex 1.4: Correlation of sectoral GDP growth Wholesale Electricity Communicati Services to Hotels and Services to General Tax to Agriculture Minning Industry Construction Transport and retail Finance Housing Bi-nationals GDP and water ons entreprises restaurants houses government products trade Agriculture 1.00 0.10 0.25 0.43 -0.02 0.67 0.01 0.64 0.09 -0.31 -0.16 0.50 -0.08 -0.38 0.46 -0.04 0.81 Mining 1.00 0.21 0.81 0.25 0.25 -0.11 0.38 0.24 -0.44 -0.10 0.26 0.50 0.27 0.22 0.08 0.33 Industry 1.00 0.30 0.30 0.15 0.19 0.44 0.28 -0.01 0.21 0.53 0.06 0.54 0.45 0.10 0.50 Construction 1.00 0.11 0.34 -0.18 0.58 0.39 -0.58 -0.06 0.42 0.47 0.22 0.30 -0.04 0.55 Electricity and water 1.00 0.03 0.21 0.41 0.29 0.18 0.26 0.47 0.03 0.34 0.61 0.21 0.30 Transport 1.00 0.09 0.62 0.09 -0.32 -0.36 0.32 0.05 -0.38 0.23 0.15 0.67 Communications 1.00 0.03 0.29 0.04 0.00 0.33 0.15 0.26 0.21 0.35 0.25 Wholesale and retail trade 1.00 0.18 -0.30 -0.11 0.54 0.13 -0.01 0.71 0.28 0.86 Finance 1.00 -0.49 0.42 0.61 0.57 0.54 0.09 0.06 0.33 Housing 1.00 -0.13 -0.20 -0.63 -0.12 0.18 0.13 -0.32 Services to entreprises 1.00 0.21 0.38 0.46 0.05 -0.31 -0.10 hotels and restaurants 1.00 0.35 0.30 0.49 0.06 0.67 Services to houses 1.00 0.43 -0.18 -0.04 0.14 General government 1.00 0.08 0.14 0.03 Tax to products 1.00 0.42 0.71 Bi-nationals 1.00 0.41 GDP 1.00 Source: Central Bank of Paraguay, World Bank staff calculations. Annex 2.1: List of interviewees and interview guide (Borda, Anichini, and Ramirez (2013)) List of interviewees Area Actor Agricultural Production Asociación Productores de Soja FECOPROOD CAPPRO: Cámara Paraguaya de Exportadores de Aceite Seeds COPATIA Relmo Paraguay Aprosemp Inputs Diagro S.A. Agrofield Equipment and Machinery COMAGRO-ROCKING Campos del Mañana Storage CAFI SILOMAQ Trasport Naviera Mercosur Multimar Technical assistance and consultrancy Agrotec Finance Banco Nacional del Fomento 70 Cooperativa Caapibary Cooperativa Ycua Bolaños Reefers Frigorifico Concepción Input providers: veterinary products, salts, minerals Ciavet LASCA Biotechnology: Embryonic reproduction Gyba SA Agricultural Insurance Sancor Garantia de Seguros. General Insurance La Agricola Seguros Generales La Consolidada Water and Electricity ESSAP Investment and Export REDIEX Think tanks CADEP CECTEC Guía de entrevistas a) ¿A qué atribuye la gran volatilidad de la producción agropecuaria/ del PIB agropecuario y como afecta a este sector? b) ¿Cómo cambian las decisiones de negocio en el sector agropecuario en anos malos y buenos del sector? c) ¿Cual es el impacto de las fluctuaciones del sector agropecuario sobre las inversiones en la economía, el consumo, el empleo, la tasa de cambio? d) ¿Cuál es la naturaleza y el nivel de relacionamiento con el sector agropecuario de su negocio? e) ¿Cuáles son la consecuencias de la volatilidad en su sector/ comercio/ empleo/ ingreso/ actividad económica? f) ¿Cómo cambian en su sector/ comercio/ empleo/ ingreso/ actividad económica las decisiones de negocio debido a las fluctuaciones en el sector agropecuario? g) ¿Por favor especifica los impactos directos y indirectos a su sector/ comercio/ empleo/ ingreso/ actividad económica de las fluctuaciones en el sector agropecuario? h) ¿Cómo y en qué grado afecta la volatilidad del PIB agropecuario a su sector? 10% ? 100%? i) En su opinión, además de su sector ¿A quién más y como les afecta la volatilidad? j) ¿Qué Riesgos y Oportunidades usted ve en la volatilidad que origina en el sector agropecuario? k) ¿Qué mecanismos existen para asegurarse en contra de/ para transferir/ para evitar estos riesgos? 71 l) ¿Usa Usted cualquier mecanismo de aseguro contra los riesgos? ¿Si no, porque no? ¿Que necesitaría cambiar para que puedan usar un aseguro? ¿Si usa mecanismos de aseguro cuales son? ¿Cómo funcionan? m) ¿Qué se puede hacer en términos de política pública y/o intervención de los agentes económicos para mitigar/evitar la volatilidad/ el impacto de la volatilidad? Annex 3.1: Traditional measures for agricultural risk management Table: Potential Risk Management Mechanisms Household/Community Markets Governments Mitigation  Sharecropping  New technology  Irrigation infrastructure  Water resource management  Improved seed  Extension  Soil drainage  Formal savings  Agricultural research  Use of resistant seeds  Weather data systems  Crop calendars  Diversification of export Risk Management Strategy  Crop diversification markets  Income/labor diversification  Diversification of agricultural  Savings in livestock production  Food buffer stocks  Farmer self-help groups Transfer  Risk pooling (peers, family  Insurance State support for insurance  members)  Hedging Derivatives or macro-level   Money lenders insurance Trading  State-sponsored hedging  Commodity exchange Coping  Sale of assets  Formal lending  Disaster relief  Migration  Risk sharing (input  Humanitarian aid suppliers, wholesalers)  Contingent financing Source: Adapted from Weather Index Insurance for Agriculture: Guidance for Development Practitioners (2011) Annex 3.2: Insurance products Table: Comparison of Insurance Products Product Summary Perils Benefits Challenges Named peril  Specific perils  Hail, fire  Simple policy  Individual farmer crop  Damage-based policy  Suited to  Limited farmer details loss assessment insurance measures percent damage localized, needed at point of sale  Loss assessment in field independently  cost in small Traditional products Transparent loss  Agreed sums insured occurring, assessment farmer systems  Typically unsubsidized sudden perils  Manageable adverse  Not suited to and run by private sector selection and moral complex perils, hazard especially drought or pest Multiple Peril  All perils  Wide list,  More easily made into a  Individual farmer Crop  Yield-based policy difficult to “universal� product loss assessment, Insurance measures harvested yield exclude risk type major loss 72 (MPCI) compared to average  Source of loss  Limited technical adjustment task, yield not identified adaptation required for impartial loss  Costly; often requires  May include diff. crops adjustment subsidy quality loss,  Farmers typically want difficult  Problematic for small price risk and understand this  Adverse selection farms insurance (worst farmers  Successful in a couple of  Indemnifies each farmer benefit) developed countries according to yield  Moral hazard (exploitation of policy)  Major work to set up yield history for each farmer, poor data  High premium and administrative cost  Not suited where farms are small Area yield  Farmers in given district  Wide list  No adverse selection,  Local perils will index all treated equally  Source of loss moral hazard, not result in insurance  MPCI but on area not identified individual farmer loss payout average yield  May include adjustment  Yield history at  Effective where similar quality loss,  Low administrative local district level exposures affect whole price risk costs often not district  Can address catastrophe available or  NAIS is largest program perils affecting group reliable (India)  Farmer enrollment easy  Basis risk at local  Index-based products Captures all causes of level depends on yield loss district area and peril Weather  Payouts based on  Rainfall  No adverse selection,  Basis risk is key Index weather station deficit and moral hazard, challenge Insurance measurement excess; individual farmer loss  Setting up index (WII)  Index trigger, exit, temperature adjustment parameters is increments set to  Basis risk  Can address catastrophe technically expected loss of yield minimized for perils affecting group complex  Can be complex to gradual  Transparent, objective  Need good design events meteorological service meteorological  Limited experience to data (MET) and agronomic date  Easier to reinsure data, crop modeling  Difficult to correlate damage for sudden-impact weather Source: Adapted from Weather Index Insurance for Agriculture: Guidance for Development Practitioners (2011) 73 Map 74